The Biology of

Musical Performance and

Performance-Related Injury

Alan H. D. Watson

THE SCARECROW PRESS, INC.

Lanham, Maryland • Toronto • Plymouth, UK
2009
SCARECROW PRESS, INC.

Published in the United States of America

by Scarecrow Press, Inc.
A wholly owned subsidiary of
The Rowman & Littlefield Publishing Group, Inc.
4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706
www.scarecrowpress.com

Estover Road
Plymouth PL6 7PY
United Kingdom

Copyright © 2009 by Alan H. D. Watson

All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, without the prior permission
of the publisher.

British Library Cataloguing in Publication Information Available

Library of Congress Cataloging-in-Publication Data

Watson, Alan H. D., 1953–
The biology of musical performance and performance-related injury / Alan H. D. Watson.
p. cm.
Includes bibliographical references and index.
ISBN-13: 978-0-8108-6358-3 (cloth : alk. paper)
ISBN-10: 0-8108-6358-8 (cloth : alk. paper)
ISBN-13: 978-0-8108-6359-0 (pbk. : alk. paper)
ISBN-10: 0-8108-6359-6 (pbk. : alk. paper)
[etc.]
1. Music—Performance—Physiological aspects. 2. Music—Performance—Psychological
aspects. I. Title.
ML3820.W27 2009
612.002'478—dc22 2008034025

⬁ ™ The paper used in this publication meets the minimum requirements of American
National Standard for Information Sciences—Permanence of Paper for Printed Library
Materials, ANSI/NISO Z39.48-1992.
Manufactured in the United States of America.
To Doris Watson, who taught me what it means
to communicate through music.
 Contents

CD Contents: Teaching Materials vii

Preface xi
Acknowledgments xvii
Reference Abbreviations xix

1 Introduction to the Tissues of the Body 1

2 Posture and the Back in Musical Performance 17
3 The Shoulder, Arm, and Hand: Structure and Problems 42
4 Breathing in Singing and Wind Playing 101
5 The Voice: Management and Problems 139
6 The Embouchure and Wind Playing 193
7 The Structure and Organization of the Brain 213
8 How the Performance of Music Affects the Brain 232
9 Hearing and the Processing of Musical Sound by the Brain 276
10 Performance-Related Stress and Its Management 332

Index 353
About the Author 369

v
 CD Contents
Teaching Materials

FIGURES

Copies of color versions of all of the figures in jpeg format.

VIDEO FILES

Video files are in Windows Media file
Windows Media Player. Apple users can
Flip4Mac Windows Media Components
center. This program or Windows Media
the Microsoft Windows download center.
(.wmv) format, which can be played on
play them on QuickTime by downloading
freeware from the QuickTime download
Player for Mac can also be obtained from

video 3-1 Tendon Linkages.wmv: “Tendon linkages in the hand” (a demonstration of

Lindburg-Comstock syndrome)
video 4-1 plethysmograph sing.wmv: “Recording chest and abdominal movements
during singing” (demonstrates the use of inductive plethysmography to record re-
spiratory movements in singers)
video 4-2 plethysmograph wind.wmv: “Recording chest and abdominal movements
during wind playing” (demonstrates the use of inductive plethysmography to record
respiratory movements in wind players)
video 4-3 Konno Mead sing.wmv: “Using Konno-Mead diagrams to understand
breathing movements during singing”
video 4-3 Konno Mead wind.wmv: “Using Konno-Mead diagrams to understand
breathing movements during wind playing”
video 5-1 normal larynx.wmv: “Movements of the vocal folds” (as seen using stro-
boscopic video laryngoscopy). Images and video clips from the Voice Clinic of the
University of Iowa Department of Otolaryngology–Head and Neck Surgery, Iowa
City; used with permission.

vii
viii CD Contents

video 5-2 singers formant.wmv: “Voice projection; the singers’ formant” (demonstra-
tion of the enhancement of vocal projection through formant manipulation, in the
mezzo-soprano voice)
video 5-3 edema nodules.wmv: “Problems with the vocal folds; Reinke’s edema and
vocal nodules.” Images and video clips from the Voice Clinic of the University of
Iowa Department of Otolaryngology–Head and Neck Surgery, Iowa City, and Clark
Rosen, M.D., University of Pittsburgh Voice Center; used with permission.

ANIMATIONS (POWERPOINT)

animation 2-1 stand-sit.ppt: “Lumbar lordosis and posture” (the effect of standing and
sitting postures on lumbar curvatures)
animation 2-2 violin posture.ppt: “The case of the twisted violinist” (postural prob-
lems arising from holding the violin between the shoulder and cheek)
animation 2-3 piano posture.ppt: “Piano posture” (the effect of seat position and
height on posture at the piano)
animation 3-1 shoulder.ppt: “Moving the shoulder joint” (the shoulder joint and its
stabilization by rotator cuff muscles)
animation 3-2 shoulder problems.ppt: “Problems with the shoulder” (bursitis and rota-
tor cuff syndrome)
animation 3-3 lumbricals.ppt: “Control of finger movement” (the role of the exten-
sor and flexor muscles that lie in the forearm, and of the lumbrical muscles of the
hand)
animation 3-4 retinacula of the wrist.ppt: “Wrist position during instrumental perfor-
mance (1)” (the role of the retinacula in containing tendon movement at the wrist)
animation 3-5 wrist position.ppt: “Wrist position during instrumental performance
(2)” (the effect of radial and ulnar deviation on tendons and nerves at the wrist)
animation 3-6 ulnar nerve.ppt: “Ulnar nerve entrapment at the elbow” (problems af-
fecting the ulnar nerve)
animation 4-1 breathing.ppt: “Breathing movements” (the role of the diaphragm and
abdominal wall muscles in breathing)
animation 4-2 breathing (continuous).ppt: (the action from the previous animation
running as a continuous loop)
animation 5-1 larynx-height.ppt: “Laryngeal movements in swallowing and yawning”
(the actions of muscles that raise and lower the larynx)
animation 5-2 larynx-breathing.ppt: “The vocal folds and breathing” (actions of the
principal muscles that open and close the glottis during breathing)
animation 5-3 larynx-voice quality.ppt: “Voice quality” (muscles controlling the lar-
ynx in whispering, breathy phonation, flow phonation, and pressed phonation)
animation 5-4 larynx-pitch.ppt: “Control of vocal pitch” (muscles and laryngeal
movements involved in changing vocal pitch)
animation 5-5 vocal fold cycle.ppt: “Movements of the vocal folds during phona-
tion” (views of vocal fold movements and the flow glottogram over the vocal fold
cycle)
 Teaching Materials ix

animation 5-6 vocal fold cycle (continuous).ppt: (as above but running as a continu-
ous loop)
animation 5-7 voice timbre.ppt: “Vocal timbre” (schematic view of the principles
behind changes in voice timbre with changing register)

SOUND PRESENTATIONS (POWERPOINT)

SP 9-1 sound identification.ppt: “Identifying the sounds of instruments” (harmonic

structure of complex sound and the role of timbre in sound identification)
SP 9-2 Tartini tones.ppt: “Tartini tones” (demonstration of the appearance of a third
pitch in the Tartini tone illusion)
SP 9-3A, B, C Missing fundamental illusions (demonstration of the properties of the
missing fundamental illusion in which virtual pitch is generated by an incomplete
harmonic series):
SP9-3A.ppt: “The missing fundamental illusion 1”
SP9-3B.ppt: “The missing fundamental illusion 2: The illusion fails at high fre-
quencies”
SP9-3C.ppt: “Pitch perception using ‘best fit’ ”

MUSIC EXTRACT

Smetana quartet Mavron.wma: An extract from Smetana’s quartet “Ma Vlast,” last
movement, Vivace, which contains a musical representation of tinnitus. Played by
the Mavron Quartet (Christiana Mavron, Katy Rowe, Rebekah Brown, and Abigail
Blackman). See the acknowledgments.
 Preface

The performance of music at the highest level requires a degree of skill that certainly
matches and arguably exceeds that of most other physical activities, yet while the
training of musicians deals extensively with the structure of music, its history, and its
interpretation, it has until recently almost entirely neglected the physical demands that
it makes on the body. The teaching of instrumental and singing technique has tradi-
tionally and indeed necessarily been based on the subjective experience of individual
teachers, sometimes resting on erroneous notions of the way in which the body works.
One of the reasons for this is that our conscious sensations do not always provide a
reliable guide to what our body is doing. As a result, it is often necessary to fall back
on metaphor or vague subjective descriptions to try to convey what is being demanded
of the pupil. This is a major barrier to passing on the knowledge gained through a life-
time of performance, as the pupil is often left confused about how an important aspect
of technique is achieved. While it must be acknowledged that objective information
of the many physical processes crucial to performance remains far from complete,
there nevertheless exists a considerable body of knowledge that would be very useful
to practicing musicians if only they could access it. Unfortunately, this is scattered
through the science literature, frequently in publications that are hard to obtain for
those outside a large academic institution. Furthermore, it is generally couched in the
arcane language of science, which will be opaque to most musicians unless they are
provided with an introduction to the world and vocabulary of biology. The object of
this book is to help performers overcome this barrier by presenting them with accurate
information on the biological principles that underlie their craft. From this beginning
they can move forward to develop a rational approach to the development of tech-
nique. It should also allow those who would like to delve deeper into the science to
access and interpret original sources and perhaps ultimately to become involved in
studies that will extend our understanding of performance biology.
Although the physical challenges faced by musicians might be similar to those of
athletes, they differ in one crucial respect: while the combination of declining physical
powers and accumulated injury leads most athletes to retire in their thirties, musicians
are expected to maintain elite levels of performance throughout a career that typically
spans many decades. It is therefore not enough for musicians to reach a high level
xi
xii Preface

of proficiency by the time they leave a music college or conservatoire. They must
be equipped with a technique that is sustainable in the long term and capable of sup-
porting the intense levels of playing demanded of the contemporary performer. If you
speak to any professional musician, he or she will tell you either from personal expe-
rience or from that of close colleagues that the profession carries a considerable risk
of a career-threatening injury. Until recently, however, health problems were rarely
discussed openly among players and managers. One reason for this was the not unjus-
tified feeling that any admission of weakness might cost the sufferer his or her hard-
won position in the highly competitive world of professional music. This is slowly
beginning to change, in no small measure through the efforts of a group of musicians
who, having suffered playing-related problems, have been moved to write about their
experiences, the lessons they have learned from them, and the strategies they have
developed both to overcome them and to avoid further difficulties. Large employ-
ers of musicians such as orchestras and opera companies are increasingly expected
to look after the health of their players. Within the UK, this has been formalized by
new endeavors such as the Healthy Orchestra Charter and the Sound Ear Initiative. Of
course as with all forms of occupational health, prevention is the most effective form
of treatment, but this requires that good practice be instilled in players from the very
earliest stages of their training. Even a cursory examination of the postures adopted
by many professional musicians on stage amply demonstrates that such an approach
is long overdue. For schemes aimed at reducing the risk of injury to performers to be
viable, however, information on which these can be based must be widely available,
and this is one of the objectives of this book.
Over the last two decades, arts medicine has emerged as a distinct discipline and
a considerable number of treatment centers for it have sprung up, particularly in the
United States and in some European countries such as Germany, the Scandinavian
countries, and more recently Spain and the UK. This has arisen from the realization
that not only do the intense physical demands made on professional musicians make
them at least as vulnerable to injury as professional athletes, but that their satisfactory
treatment also requires a similarly specialized approach. The growth of this discipline
has undoubtedly been aided by the fact that many clinicians have an interest in music.
Finding practitioners with the necessary skills and insights to provide rapid and accu-
rate diagnosis and treatment, however, still remains something of a lottery. It is difficult
for health professionals such as doctors or physiotherapists to develop expertise in the
treatment of a particular type of patient if they see only one or two a year. Many of the
problems suffered by musicians are well known to medicine, but for effective treat-
ment it is vital that these are viewed in their vocational context. From this perspective
many aspects that might initially appear perplexing quickly become clear and we can
progress from treating the symptoms to tackling the root cause. It is also important to
realize that the level of function that must be restored in an injured musician to allow
a return to even a basic level of professional performance is much greater than that
which would be acceptable for most other patients presenting with similar problems.
Furthermore, the treatment may have to be modified in light of the particular instru-
ment played, similar to how athletes are rehabilitated differently based on their sports.
 Preface xiii

The information presented in this book may therefore be of value to health profession-
als who find themselves treating injured musicians for the first time.
Fortunately, in some countries there now exist national organizations that actively
foster links between medicine and the arts, for example, Médecine des Arts (France),
the Deutsche Gesellschaft für Musikphysiologie und Musikermedezin (Germany),
and the Performing Arts Medicine Association (PAMA) in the U.S. These each take
their own approach, but in the UK the British Association of Performing Arts Medi-
cine (BAPAM—BAPAM.org.uk) has set up regional clinics for the initial diagnosis
of performance-related health problems (in dancers and actors as well as musicians),
from which performers can be referred to doctors and other health professionals who
are specialists in their treatment and rehabilitation. Together with the Musicians Be-
nevolent Fund, BAPAM is also increasingly involved in health promotion and injury
prevention.
An area of research that has emerged even more recently than arts medicine is that
of performance biology—the objective study of the physical and mental processes that
underlie singing, instrumental performance, and dance. This field has the ultimate aim
of putting musical performance training on a rational basis so that not only are the
most effective strategies used, but the chance of developing an injury is minimized.
Early investigations in the field were often not rigorous enough and suffered from
small numbers of study subjects, but this is slowly beginning to change. Notable cen-
ters for the study of music biology, psychology, and medicine are now found in such
places as the Institute for Music Physiology and Musicians’ Medicine in Hannover,
Germany, the University of Montreal, Harvard Medical School, and the Royal Col-
lege of Music in London. In addition, many researchers have made contributions to
music biology as a sideline from their main research. The field is currently in the state
that sports biology reached several decades ago, but some music teachers are already
trying to apply the available information to the teaching and treatment of musicians.
Many institutions have set up courses covering at least some areas of performance
biology and medicine; however, the lack of a suitable textbook is a major impediment
to the wider development of such courses, which should form part of the core curricu-
lum of all music schools. This book is intended to remedy this situation by providing
a broad introduction to the major areas of human biology of relevance to performers.
One often feels, when reading the literature of music biology, that although musi-
cians are interesting subjects for anatomists, physiologists, and neuroscientists, they
are by and large passive participants in many studies and don’t always receive the
return they deserve for their participation. The science of performance has a great deal
to offer performers and it has been my experience in Cardiff that if they are interested
in these matters, musicians can quickly assimilate sufficient knowledge to take an
active part in and ultimately direct such studies for their own ends. Another object of
the book is therefore to help facilitate this transition. It is not simply an account of
human anatomy or physiology to which have been added a few examples of medical
problems found in musicians. Each chapter is written specifically from the viewpoint
of musical performance and makes heavy use of relatively unknown literature sources
on music biology, which provide important information that is not to be found in
xiv Preface

standard biology or medical textbooks. For example, you will not find a discussion
of the limitations to individual finger movement even in the definitive text of Gray’s
Anatomy, and no medical textbook provides an account of current research into respi-
ratory strategies of relevance to singers or discusses theories of the control of register
or voice projection. With this book I have endeavored, wherever possible, to address
questions that musicians need answers to and to point out some of the limitations of
current knowledge. I have also sought to assess the validity of some key notions that
are commonly held by musicians, although I will not pretend that I can provide the
last word on any of these issues. Rather, the position presented here is a starting point
from which I hope musicians themselves will carry the baton onward.
This book originated from a series of lectures given at Cardiff University and the
Royal Welsh College of Music and Drama. Preparing these lectures brought home to
me the difficulty for musicians (and indeed biologists) of finding accessible sources
of information on music biology. For musicians this is exacerbated by having to deal
with the language of biology, and the book tries to assume no prior knowledge of this.
Even for biologists, some of the most active areas, such as the study of what goes
on in the brain when we listen to music, can be quite impenetrable. This is not only
because of the terminology used, but also because of the difficulty of trying to un-
tangle different methods and approaches, which can generate results that may at first
appear conflicting or confusing. Here then is a starting point, but keep in mind that
this field is still in its infancy; it will be clear to those who read the book that many
concepts of central importance are in urgent need of comprehensive reevaluation and
study. The ease with which physiological data can now be collected, digitized, and
analyzed with modest computing resources means that a great deal can be achieved
with relatively inexpensive equipment if directed at a well-defined and manageable
objective. However, it is all too easy to collect large amounts of data from which it is
difficult to extract vocationally meaningful conclusions. It is therefore essential that
experienced professional musicians, as well as those with a background in science or
medicine, be involved at all stages of such studies, from setting the initial objectives,
through experimental design and execution, to analysis and interpretation. If this book
encourages just a few performers in taking on such a role, it will have made an im-
portant contribution to ensuring that musicians will in the future have a solid body of
information on which to build their technique.

HOW TO USE THIS BOOK

For some readers, it will be sufficient that the book serves as a source of reference
or of background information on the biology of technique. For others it will form a
bridge that will provide access to the original literature on music biology as it cur-
rently stands and as it evolves. Each chapter therefore ends with a list of references of
the main sources of information on which it is based. The chapters’ differing lengths
reflect to some extent the depth of our understanding in each area. In the hope that the
book will encourage the development of courses in music biology in music colleges
and conservatoires, I have produced a set of teaching aids, which can be found on the
 Preface xv

accompanying CD. One element is a set of simple illustrations that are intended to
present key anatomical information in a clear and approachable way. You will find
them printed in black and white in the text, but on the disc most are produced in color,
which will make them more engaging and perhaps easier to interpret. Written descrip-
tions of movements and postures can sometimes be difficult to understand, though
they are easily appreciated when seen. The disc therefore includes a set of animations
that illustrate some key concepts of posture, movement, and laryngeal function. These
are in the form of PowerPoint presentations but can be run even if you do not have the
program by clicking on pptview.exe in the appropriate directory. You can then select
the animation you require, whereupon it will open fullscreen and run automatically
with preset timings. Alternatively, if you have PowerPoint, you can open them with
the program and manipulate or disable the timings. One of the disappointing aspects
of reading accounts of auditory illusions—important for our appreciation of music—is
that they can’t be heard unless you have access to facilities for generating or manipu-
lating sound waveforms. The CD therefore also contains a number of PowerPoint files
(labeled Sound Presentations) that reproduce the auditory effects discussed in chapter
9. Once the files are opened fullscreen, some will run automatically while others will
produce the sounds only when the loudspeaker symbols are clicked. For some, you
will also need to use the Page Down key to go to the next screen. If you are using them
in a lecture hall, the clarity of the illusions will depend on the quality of the sound sys-
tem, though for private listening even an inexpensive set of headphones is adequate.
The disc also contains an excerpt from Smetana’s Quartet “Ma Vlast” (From My
Life), which gives a musical rendition of a hearing problem known as tinnitus. This
is played by the Mavron Quartet (see the acknowledgments). Finally, I have created
a number of short videos, mainly dealing with breathing for singers and wind players
and with aspects of laryngeal function.
 Acknowledgments

I am most grateful for the generosity of Roger Chaffin, Robert Harrison, Michael
Karnell, Clark Rosen, Johan Sundberg, and Ingo Titze for giving me permission to re-
produce images from their work; to Buddug Verona James for discussions on singing
technique and for her involvement in the respiratory recordings shown in the book and
heard on the disc; to Robert Santer, who was involved in creating the video of tendon
anomalies; and to the Mavron Quartet (Christiana Mavron and Katy Rowe [violins],
Rebekah Brown [viola], and Abigail Blackman [cello], www.mavronquartet.co.uk)
for making the recording of the Smetana Quartet that appears on the CD. Louise
Atkins of the Royal Welsh College of Music and Drama (RWCMD) played a major
role in collecting and analyzing the respiratory data from wind players that form the
basis of some of the figures and videos. This is part of a project to raise awareness in
musicians of the physical aspects of performance and has been supported by a grant
for public engagement in science from the Wellcome Trust, held by Kevin Price and
me. I would also like to acknowledge Zoe Smith, who has encouraged the integration
of the material that forms this book into the music curriculum of the RWCMD, and
Sanchita Farruque of the British Association of Performing Arts Medicine, who was
brave enough to try some of the support materials on a variety of audiences. Finally,
I am greatly indebted to Kevin Price (head of brass teaching at RWCMD) for acting
as a sounding board for many of the ideas presented here, for liaising with performers
who have contributed to the production of some of the illustrative material, and for
helping to ensure that the treatment of the material is vocationally relevant to work-
ing musicians. The fact that this project has come to fruition owes a great deal to his
support and enthusiasm.

xvii
 Reference Abbreviations

Below is a list of abbreviations used in the references at the end of each chapter and
the corresponding full name for each publication.

Acta Anat Basel (Basel) Acta Anatomica (Basel)

Acta Otolaryngol Acta Otolaryngologica
Adv Cog Psychol Advances in Cognitive Psychology
Adv Neurol Advances in Neurology
Am J Med American Journal of Medicine
Am J Orthod American Journal of Orthdontics
Am J Otol American Journal of Otolaryngology
Am J Physiol American Journal of Physiology
Am J Psychiatry American Journal of Psychiatry
Am J Psychol American Journal of Psychology
Anat Rec Anatomical Record
Anat Rec A Discov Mol Anatomical Record A: Discoveries in Molecular,
Cell Evol Biol Cellular, and Evolutionary Biology
Angle Orthod Angle Orthodontist
Ann Allergy Annals of Allergy
Ann NY Acad Sci Annals of the New York Academy of Sciences
Ann Otol Rhinol Laryngol Annals of Otology, Rhinology and Laryngology
Annu Rev Psychol Annual Review of Psychology
Arch Environ Health Archives of Environmental Health
Arch Med Res Archives of Medical Research
Arch Néerl Phon Expér Archives Néerlandaises de Phonétique Expérimentale
Arch Otorhinolaryngol Archives of Otorhinolaryngology
Arch Phys Med Rehabil Archives of Physical Medicine and Rehabilitation
Aust Dent J Australian Dental Journal
Behav Brain Res Behavioral Brain Research
Behav Res Ther Behaviour Research and Therapy
Biofeedback Self Regul Biofeedback and Self Regulation

xix
xx Reference Abbreviations

Biol Psychol Biological Psychology

BMC Dermatol Biomed Central Dermatology
BMC Neurosci Biomed Central Neuroscience
Brain Cogn Brain and Cognition
Brain Res Brain Res Rev Brain Research: Brain Research Reviews
Brain Res Cogn Brain Res Brain Research: Cognitive Brain Research
Brit J Psychol British Journal of Psychology
Br Dent J British Dental Journal
Cereb Cortex Cerebral Cortex
Clev Clin Q Cleveland Clinic Quarterly
Clin Anat Clinical Anatomy
Clin Biomech (Bristol, Clinical Biomechanics (Bristol, Avon)
Avon)
Clin Neurphysiol Clinical Neurophysiology
Clin Sci (Lond) Clinical Science (London)
CMAJ Canadian Medical Association Journal
Cogn Brain Res Cognitive Brain Research
Curr Opin Neurobiol Current Opinion in Neurobiology
Ear Hear Ear and Hearing
Epilepsy Behav Epilepsy and Behavior
Eur J Anat European Journal of Anatomy
Eur J Neurosci European Journal of Neuroscience
Eur J phys European Journal of Physiology
Eur Respir J European Respiratory Journal
Exp Brain Res Experimental Brain Research
Folia Phoniatr (Basel) Folia Phoniatrica (Basel)
Forsch Komplementarmed Forschende Komplementärmedizin und Klassiche
Klass Naturheilkd Naturheilkunde
Hand Clin Hand Clinics
Hum Brain Mapp Human Brain Mapping
Hum Mov Sci Human Movement Science
Int J Dermatol International Journal of Dermatology
Int J Pediatr International Journal of Pediatric
Otorhinolaryngol Otorhinolaryngology
Int J Ther Rehab International Journal of Therapy and Rehabilitation
J Acoust Soc Am Journal of the Acoustical Society of America
JAMA Journal of the American Medical Association
J Am Dent Assoc Journal of the American Dental Association
J Anat Journal of Anatomy
J Appl Physiol Journal of Applied Physiology
J Asthma Journal of Asthma
J Biomech Journal of Biomechanics
J Bone Joint Surg Am Journal of Bone and Joint Surgery (American volume)
J Clin Neurosci Journal of Clinical Neuroscience
 Reference Abbreviations xxi

J Cogn Neurosci Journal of Cognitive Neuroscience

J Consult Clin Psychol Journal of Consulting and Clinical Psychology
J Hand Surg (Am) Journal of Hand Surgery (American volume)
J Hand Surg (Br) Journal of Hand Surgery (British volume)
J Hand Ther Journal of Hand Therapy
J Int Double Reed Soc Journal of the International Double Reed Society
J Manipulative Physiol Journal of Manipulative and Physiological
Ther Therapeutics
J Music Theory Journal of Music Theory
J Neurol Sci Journal of Neurological Science
J Neurophysiol Journal of Neurophysiology
J Neuroradiol Journal of Neuradiology
J Neurosci Journal of Neuroscience
J Res Music Educ Journal of Research in Music Education
J R Soc Med Journal of the Royal Society of Medicine
J Speech Hear Res Journal of Speech and Hearing Research
J Voice Journal of Voice
Learn Mem Learning and Memory
Logoped Phoniatr Vocol Logopedics, Phoniatrics, Vocology
Md Med J Maryland Medical Journal
Med Hypotheses Medical Hypotheses
Med J Aust Medical Journal of Australia
Med Prob Perform Artists Medical Problems in Performing Artists
Mem Cognit Memory and Cognition
Mov Disord Movement Disorders
Muscle Nerve Muscle and Nerve
Nat Neurosci Nature Neuroscience
Nat Rev Neurosci Nature Reviews Neuroscience
Neuroendocrinol Lett Neuroendocrinology Letters
Neuroreport Neuroreport
Neurosci Lett Neuroscience Letters
Neurosci Res Neuroscience Research
Obstet Gynecol Obstetrics and Gynecology
Occup Med (Lond) Occupational Medicine (London)
Otolaryngol Clin North Am Otolaryngologic Clinics of North America
Otolaryngol Head Neck Otolaryngology: Head and Neck Surgery
Surg
Plast Reconst Surg Plastic and Reconstructive Surgery
Proc Natl Acad Sci USA Proceedings of the National Academy of Sciences (U.S.)
Psychol Music Psychology of Music
Psychol Sci Psychological Science
Respir Physiol Respiratory Physiology
Sci Am Scientific American
Semin Neurol Seminars in Neurology
xxii Reference Abbreviations

Speech Hear Res Speech and Hearing Research

Trans Med Soc Lond Transactions of the Medical Society of London
Trends Cogn Sci Trends in Cognitive Sciences
Trends Neurosci Trends in Neuroscience
 Chapter One

Introduction to the
Tissues of the Body

Before we examine the biological principles that underlie the performance of music, it
is first necessary to know something about the structure of the basic tissues that make
up the body. The objective of this chapter is therefore to provide a brief introduction
to this topic; taking the time to go through it carefully will make it easier to understand
the following chapters. It is important to recognize, however, that this is not intended
to provide a comprehensive overview of human body structure, but is highly focused
on subjects that are central to the rest of the book. It does not, for example, deal with
the circulatory system or with the major organ systems of the body. Even the subjects
that are covered are dealt with on a need-to-know basis, in order to get readers who are
musicians into the sections of the text that concern them as quickly as possible. If they
find the biology interesting, some may wish to delve more deeply into conventional
textbooks at a later date, and this will undoubtedly add to their appreciation of the is-
sues presented here. However, such books generally have little to say on the specific
aspects of biology that are central to musical performance.
This chapter deals with the basic elements of the body excluding the organs, that
is, bone, cartilage, and joints; muscles and tendons; the skin; and the organization of
the spinal cord. It deals with sensation and in particular with the nature of pain. The
section on skin will also address some matters of particular relevance to musicians in
the form of a discussion of contact dermatitis. The succeeding chapters will build on
what is presented here and provide additional basic information as necessary for the
understanding of performance-related function and injury. Information is introduced
where its relevance for performers is best appreciated, so it should not be necessary
to keep flipping from one chapter to another to get the complete picture. For example,
the section on tendon sheaths will be found close to where the problems associated
with tendons are discussed. An account of the respiratory system is left entirely to the
chapter on breathing strategies, and an overview of the organization of the brain is
presented in the chapter that immediately precedes those dealing with music and the
brain, the auditory system, and stress, all of which are dependent on the foundation
this chapter provides.

1
2 Chapter One

BONE, CARTILAGE, AND JOINTS

When you think of bone, what probably comes to mind is the dry skeleton and with
it the idea that bone is an inert and unchanging structural material like concrete, but
this is far from the truth. Both the calcium and the phosphates that make up the hard
material in bone are important chemicals used by all cells in the body for a variety of
purposes. While a major function of bone is to provide structural support for the body,
it also acts as a reserve supply of these materials. Bone also responds to the demands
that are made upon it and will slowly increase in mass if great forces are applied to it.
Historically this is seen, for example, in English longbowmen, whose bows required
a force of up to 65 kilograms to draw. Examination of skeletons of archers recovered
from the Tudor warship the Mary Rose reveals that the bones of the load-bearing left
arm show significantly greater development than those of the drawing arm. A similar
phenomenon is seen today in tennis players, in whom the mineral content of the bones
of the racket arm may be up to 20 percent greater than in the other.
On the other hand, bone mass can also be lost. This is known as osteoporosis and
is common in elderly people of both sexes, but particularly in women, in whom it is a
consequence of the hormonal changes that take place at menopause. This reduction in
bone mass makes one more vulnerable to hip fractures and spinal deformity. In life,
bone is therefore an active tissue. Far from being dry, it not only contains cells that
regulate its structure but also a good blood and nerve supply to maintain them. As a
result, breaking a bone is a very painful experience. The structure of large bones is
designed to provide maximum strength with a minimum of weight. The central sec-
tion of the long bones of the arms and legs is made up of a cylinder of dense bone
surrounding a hollow core. The core contains marrow, a tissue that is responsible for
the production of red blood cells, the carriers of oxygen, and white blood cells, which
fight disease. Toward the ends of the long bones, the marrow is replaced by spongy
bone, which, though light, is capable of supporting considerable force.
During embryological and early postnatal development, bone develops from a
more flexible material called cartilage, which provides a matrix for the deposition of
calcium and phosphate. There are many other forms of cartilage in the body, how-
ever. Cartilage forms the ligaments that hold bones in position and the tendons that
transmit the forces generated by muscles to the bones. Cartilage also contributes to
several important components that in turn contribute to the structure of joints. We
will see in a later chapter that the skeleton of the larynx is constructed entirely from
cartilage, though bone may be deposited there later in life. Cartilage is created by cells
that secrete a variety of structural fibers with differing properties. Some of these are
elastic, while others such as collagen are thicker and less extensible. These fibers give
the cartilage its tensile strength. The gaps between the fibers are filled with space-
filling molecules, many of which help the tissue to retain water. Together these create
a gel in which the fibers are embedded. This gives the cartilage great stiffness and
a resistance to compression. The structural advantages of materials that combine a
meshwork of fibers within an incompressible matrix are reflected in their widespread
use in engineering, from prestressed concrete and fiberglass to the carbon fiber and
other composites used by the aeronautics industry.
 Introduction to the Tissues of the Body 3

Figure 1-1. A. The structure of a synovial joint. The surfaces of the bone that make contact
in the joint are covered with smooth, hard-wearing hyaline cartilage. The joint is enclosed in a
capsule, which is usually reinforced by ligaments and contains lubricating synovial fluid. B. The
relationship between muscle and tendon is shown in this diagram of a hamstring muscle, which
runs from the pelvis to the bones of the lower part of the leg. A thin strip of tendinous tissue
(cartilage) attaches the muscle to the hip (its origin). At the other end, a long, thin tendon (the
“hamstring”) emerges from the muscle and runs down the side of the knee to its attachment site
on the side of the lower leg (its insertion).

Where bones come together, they are connected by joints. Some of these allow little
to no movement, such as the ones between the bones of the skull, but most are present
to permit some degree of movement, that is, they allow the bones to articulate with
one another. The joints that are most important for the purposes of this book are what
are called synovial joints, and we will here concern ourselves exclusively with these
(fig. 1-1A). The surface of normal bone is covered with a fine membrane called the
periosteum, which is not sufficiently robust to resist the friction and wear at the con-
tact surfaces of a moving joint. These surfaces are instead covered with a special form
of cartilage called articular or hyaline cartilage. This provides a smooth, glassy surface
that is resistant to wear and is self-repairing, which is important as these moving joints
must remain intact for a lifetime. If the bony surface of the joint is shallow (e.g., the
shoulder joint), a rim made up of a different type of cartilage must be laid down to
give it added depth. In order to reduce the friction between the moving surfaces, they
are lubricated with a watery fluid (synovial fluid) produced by membranes (synovial
membranes) that lie within the joint. To stop the fluid from draining into the surround-
ing tissues, each joint is enclosed in a sac called the joint capsule. The joint is usually
reinforced with ligaments that run between the bones to keep the two sides in close
proximity and to limit the degree of movement in certain planes. The ligaments, which
are also made of cartilage, are often fused with the capsule. In some cases (such as the
jaw joint), a disk of cartilage is also interposed between the two joint surfaces.
4 Chapter One

The possible range of movement at a joint depends on the shape or the articular
surfaces. The elbow joint and the last two joints in the fingers are known as hinge
joints because they allow movement in only one plane. Joints such as those at the
wrist and at the base of the finger and thumb allow limited movement in two planes,
while the shoulder and hip joints are ball-and-socket joints that allow a much greater
range of movement.

MUSCLES AND TENDONS

The joints between the bones are moved by what is known as skeletal muscle. Such
muscle is composed of hundreds of thousands of long muscle fibers, each of which is
capable of contracting under the control of the nervous system. The fibers are rather
strange cells as each is formed from the fusion of many smaller cells. Most cells have
only a single nucleus, which contains the genetic blueprint of the whole animal, but
from which is translated the parts required by the particular cell type. Because of how
they are formed, muscle fibers each contain many nuclei. Muscle fibers, even within
the same muscle, may have different properties. Some are known as fast-twitch fibers
because they contract rapidly, but these are easily fatigued. Slow muscle fibers by
contrast not only do not contract so quickly, but can remain contracted for longer
periods without becoming exhausted. The balance between slow and fast muscle fi-
bers depends on what the muscle is used for, how it has been trained, and the genetic
makeup of the individual. Muscles have a variety of forms depending on their role
in the body. Some, such as those in the abdominal wall, are flat sheets, while others
such as the biceps are more rounded in cross-section. We generally think of muscles
as being anchored to a fixed bone at one end and a moving one at the other. Under
these circumstances, the fixed end of the muscle is described as its origin and the
moving end its insertion. It is not unusual, however, to find that the bone at either
end may be moved depending on the posture adopted. For example, a muscle on the
front of the thigh may raise the leg when we walk but pull the trunk forward at the
waist when we sit up from a recumbent position. Similarly, a muscle at the back of
the thigh may bend the knee if the weight of the body is supported by the other leg
or rotate the pelvis backward if the foot is planted firmly on the ground (fig. 1-1B).
Although many muscles are attached to bone at both ends, others insert into sheets of
fibrous connective tissue (e.g., abdominal muscles) or even the skin or other muscles
(e.g., the muscles of the face and embouchure). Where the attachments are made to
bone, some tendinous tissue forms the connection. This may be difficult to see at the
origin, where the muscle often appears to spring directly from the bone. The tendon at
the other end is usually elongated into a strap or sheetlike structure (fig. 1-1B). Long
tendons can be used to improve the mechanical advantage of the muscle, but they also
allow the muscle to lie some distance away from the joint from which it acts. In the
case of the hand, the major power muscles that flex and extend the fingers lie in the
forearm. They can therefore be much larger and stronger than would be possible if
they all lay within the hand itself, which contains much smaller muscles that provide
the fine control necessary for dextrous finger movements.
 Introduction to the Tissues of the Body 5

There are two other major classes of muscle. One of these is exclusive to the heart.
Cardiac muscle requires quite different properties from skeletal muscle. First, as the
heart beats throughout life it must be very resistant to fatigue. It must also provide
a conductive path for waves of electrical excitation to spread across the heart. Many
cardiac muscle fibers will contract rhythmically even in the absence of signals from
the nervous system. The other major type of muscle is smooth muscle. This is found
mainly in the hollow organs such as the gut and bladder, and lining the airways of the
lungs. It is also present in the walls of the arteries. This smooth muscle directs blood
to different places as the need arises by regulating the width of different arteries. For
example, when we run, the muscular walls of the arteries supplying the muscles relax
to increase their access to oxygen and nutrients. More blood is also directed through
the arteries running in the skin to dissipate the heat that the muscles produce. By
contrast, when we eat a heavy meal, the smooth muscle in the arteries in the muscles
and skin contracts to make them narrower, while the arteries supplying the gut and the
liver increase in diameter to support digestion.

SKIN AND EPITHELIA

Epithelia are sheets of cells that cover the external and internal surfaces of the body,
forming a barrier between the tissues of the body and the external environment or
the contents of organs such as the gut or bladder, but they may also be modified for
other roles such as secretion of fluids, mucus, or other compounds. For the purposes
of this book we need consider only two types of epithelia. The first covers the outer
surface of the body and we call this the skin (fig. 1-2). The second, which we will see
in chapter 5, lines the back of the nose, the throat, and the larynx.
The skin has several major functions. One is to provide a covering for the body
that is both resistant to abrasion and a barrier to the uncontrolled loss of water. The
epithelium, which forms the outer region or epidermis or outer layer of the skin, is
composed of many layers of flattened cells, which are constantly lost from the surface
and replaced from below (fig. 1-2A). It is a somewhat nauseating truth that the mate-
rial that collects on the air filters of the London Underground is mostly composed
of skin cells! Only the cells in the inner layers are alive, but these are tightly bound
together to prevent water in the underlying tissues from escaping to the surface. The
cells on the outer layers are dead, but they provide a tough surface that is resistant to
damage due to rubbing or scraping. The cells are modified to resist abrasion by being
desiccated and by containing granules of keratin, a tough protein that also makes up
fingernails and hair.
The epidermis does not have a blood supply, but the dermis (the layer of the skin
that lies directly beneath it) is well supplied by blood vessels. When the body becomes
overheated, the blood vessels expand so that heat can be more efficiently lost by ra-
diation from the skin surface. Heat is also lost by the evaporation of sweat, which is
released from sweat glands that are found in the dermis but which send ducts all the
way up to the skin surface. Most of the body is covered with hair growing from hair
follicles that are modifications of the epidermis that push down to its border with the
6 Chapter One

Figure 1-2. A. The structure of the skin. The outer layer of the skin (epidermis) is formed by a
multilayered epithelium of flattened cells. The outer cells are dead and are gradually sloughed off
to be replaced by those beneath, which are continually being produced by the innermost layer.
Beneath the epidermis lies the dermis, which contains the sweat glands and hair follicles. Small
muscles attach to the walls of the hair follicle; these raise the hair if we are cold or frightened. B.
Sensory nerve endings in the skin. The sensory nerves that supply the skin carry a variety of types
of sensation, such as pain. Some are known as naked endings in which the branches at the end of
the axons are not structurally modified. Other endings are associated with cellular structures that
have an effect on their sensitivity to different types of stimuli. Some endings lie below the dermis,
while others lie more superficially at the boundary between the dermis and epidermis. The more
superficial endings are responsible for the finest forms of tactile discrimination.

dermis, where the cells that produce the hair get their blood supply. Sebaceous glands
also empty their contents into the hair follicles, producing a thin grease that flows onto
the hair and the skin surface, which helps to make the skin more water resistant and so
prevents it from drying excessively, which could cause it to crack. Each hair follicle
is attached to a small muscle that can pull the hair erect (fig. 1-2A). This originally
evolved as a mechanism to conserve heat; though the hair on our skin is generally too
fine and sparse for this to have much effect, the response still occurs, creating what
we call “goose bumps.” Each of these represents the effects of a muscle attached to a
single hair follicle. The hairs also stand up when we are frightened or excited, causing
sensations that we may experience as “creeping,” “hair rising on the back of the neck,”
or “shivers down the spine,” depending on the context. This sometimes happens when
we have a strong emotional reaction to a piece of music.
Like many other body tissues, the properties of the skin change depending on how
it is used. Where it experiences great friction or pressure it becomes thicker and may
form calluses. This is important for guitarists and players of bowed string instru-
ments, as the calluses that build up on the pads of the fingers allow high pressures to
be exerted to keep the string in contact with the fret or fingerboard without causing
pain or blister formation. Playing an instrument can cause the skin to become modi-
fied or damaged in less helpful ways, however. In violin and viola players, pressure
and abrasion can lead to changes in the texture and color of the skin on the side of the
neck where it makes contact with the instrument. Particularly when combined with
poor hygiene, this may also lead to acne or infection. The altered skin is commonly
known as “fiddler’s neck” and ranges from a mild thickening or discoloration to dry,
 Introduction to the Tissues of the Body 7

itchy, or infected skin (Blum and Ritter 1990; Onder et al. 1999; Harvell and Maibach
1992). Occasionally, swelling that initially appears to be due to skin thickening is the
result of cysts in the salivary glands, which lie beneath the jawline or on the cheek,
where they can be compressed by the instrument. Allergic or toxic reactions to com-
pounds associated with the instrument can also contribute to the symptoms of fiddler’s
neck (Gambichler, Boms, and Freitag 2004). Some string players develop allergies
to rosin (colophony), a compound derived from the sap of pine and spruce trees, or
components of instrument varnish such as propolis. Propolis, also known as bee glue,
is a resin collected from the buds and bark of trees such as the poplar and used to seal
gaps in the hive. In some individuals, contact with these materials produces dermati-
tis (a general term indicating a swelling of the tissues in the skin) or an eczema-like
reaction. Many exotic hardwoods such as rosewoods, cocobolo, and ebony, which are
used in recorders and in high-quality clarinets and oboes, can also induce allergic skin
responses. Some chin rests may also be carved from rosewoods. Allergic reactions
are not only seen in players, but can also pose a significant problem for instrument
makers, as the dust from these woods quite often triggers asthmalike symptoms. The
chemicals in the wood that are responsible are not unique to a single species of tree, so
some experimentation may be required to find a suitable alternative for an offending
chin rest or woodwind instrument. Boxwood is generally a safe replacement as it does
not contain significant quantities of known allergens. Problems can also arise from
stains that have been used on wood. For example, paraphenylenediamine, a stain used
to give the appearance of ebony to paler and more common woods, is a well-known
allergen. As a result, chin rests and fingerboards that appear to be ebony but are not
can be a source of contact dermatitis on the neck or fingers.
If the symptoms of fiddler’s neck are caused mainly by the pressure of the instru-
ment on the skin, a number of approaches may be used to reduce them (Harvell and
Maibach 1992; Blum and Ritter 1990). Attention to hygiene and ensuring that any
cloths used over the chin rest are always clean are obvious measures. Men may find
it beneficial to grow a beard, though occasionally this causes its own problems if the
hair is curly and becomes ingrown. Changing the setup to increase the load-bearing
area of the skin surface may also offer some benefit. It has been suggested that one
way of achieving this is to move the chin rest toward the midline of the instrument
(Blum and Ritter 1990); however, one objective study of different chin rests found that
higher pressures were exerted with this type of rest than with two others whose contact
surfaces were positioned more to one side (Okner, Kernozek, and Wade 1997).
Contact with certain metals can also cause skin problems. Nickel is particularly
well known for this, though other metals such as chromium have also been implicated
to a lesser extent. Nickel is a component of alloys used in the manufacture of many
brass instruments and their mouthpieces, inexpensive flutes, and the fittings of other
woodwinds. It is frequently employed in the form of nickel silver, which is actually an
alloy of copper, zinc, and nickel. However, the situation is complicated by the fact that
the alloy that forms the bulk of the instrument may be covered by a plating of another
metal. Heat, abrasion, and interactions with sweat or saliva all conspire to increase the
risk of contact dermatitis with metals. Indeed, the flaking of skin around the mouth
(cheilitis) in brass players or flutists is sometimes due solely to physical abrasion of
8 Chapter One

the thin, wet skin around the lip (Thomas, Rueff, and Przybilla 2000). Where a reac-
tion to metal is the cause, a thin coating of a more inert metal such as silver or gold
may provide a solution. Alternatively, a mouthpiece with a plastic rim or one made
entirely of plastic can be used. Nickel is sometimes present in the metal elements of
violin and viola chin and shoulder rests and so can contribute to the development of
fiddler’s neck. It is used as a coating for steel guitar strings and in the overwound
strings of many instruments. If this irritates the skin of the fingers, an alternative
choice of string is generally available; for example, bronze or aluminum is often used
in overwindings. In electric guitars, however, almost all available strings have a major
nickel component.

THE CENTRAL AND PERIPHERAL NERVOUS SYSTEM

For the purposes of description, the nervous system is usually divided into the central
nervous system, which is made up of the brain and spinal cord, and the peripheral
nervous system, which comprises the nerves that supply the skin, muscles, joints, and
organs. The structure of the brain and its role in musical performance are dealt with
in chapters 7 and 8. Here we will confine our considerations to the spinal cord and the
peripheral nervous system.

The Spinal Cord

The vertebrae that form the skeleton of the spine each incorporate a hollow space
for the spinal cord. The cord is therefore surrounded by bone, so each vertebra must
have a pair of openings through which nerves pass to reach the rest of the body (see
chapter 2). These nerves are called the segmental nerves and they supply the muscles,
the skin, and the organs. The region of skin innervated by each segmental nerve is
called a dermatome, and in the thorax (chest region), these can be viewed as horizontal
strips on the surface of the skin. The pattern is more complicated in the cervical (neck)
region and in the lumbosacral (lower back) region, where the segmental nerves must
supply the limbs as well as the trunk. In these regions the nerves from several spinal
segments fuse together and then branch again in complex patterns to form a network
or plexus. In the case of the arm this is called the brachial plexus and for the leg, the
lumbosacral plexus (see chapters 2 and 3). We will see later that compression of the
branches of the brachial plexus can cause problems in musicians, for example, where
they emerge from the vertebral column, or where they pass between the muscles of the
neck or through a narrow space between the collarbone and the rib cage.
If we were to peer into the narrow tunnel that runs down the center of the ver-
tebral column, we would see that each segmental nerve is formed by the fusion of
many small rootlets that arise as a continuous series from both the dorsal and ventral
surfaces of the spinal cord (fig. 1-3A). At each spinal level, the dorsal bundles fuse
to form a single dorsal root and the ventral bundles, a ventral root. The dorsal and
ventral roots then join to create the segmental nerve. The section of cord giving rise to
each segmental nerve is regarded as a functional unit and is named after the vertebrae
 Introduction to the Tissues of the Body 9

Figure 1-3. The organization of the spinal cord. There are three types of nerve cells in the spinal
cord: sensory neurons, motor neurons, which supply the muscles, and interneurons, which make
connections between other neurons. Each neuron has a cell body, which contains the nucleus,
and an axon that connects it to other neurons. Motor neurons and interneurons also have many
dendrites that sprout from the cell body and lie within the gray matter, while the surrounding
white matter carries axons up and down the cord. The gray matter is divided into the dorsal horn,
which is mainly concerned with processing sensory information, and the ventral horn, which
contains the motor neurons. A. A sensory neuron that carries information from a muscle runs into
the ventral horn to contact motor neurons, though it also interacts with interneurons in the dorsal
horn. B. A sensory neuron comes from the skin and ends in the dorsal horn on an interneuron that
carries sensory information up the cord to the brain.

through which the nerve passes. However, the spinal cord is much shorter than the
vertebral column. It terminates in the midlumbar region (in the small of the back), and
the segmental nerves of the lower segments of the cord must run for some distance
down the central tunnel within the spine so that they can emerge at the appropriate
vertebral level. These abundant downward running roots are called the cauda equina
because of their superficial resemblance to a horse’s tail.
Seen in cross-section, the spinal cord is round or oval, but it has a central butterfly-
shaped area which, because of its color, is called the gray matter (fig. 1-3B). This is
surrounded by a border of paler tissue known as white matter. Regions of gray and
white matter are found throughout the central nervous system; in order to understand
their significance we need to consider the nature and diversity of nerve cells (more
details of this can be found in chapter 8). Figure 1-3 shows the structure of several
types of nerve cells (or neurons). Each has a central spherical region called the cell
body from which arise a number of short branches and one long branch. The cell body
contains the nucleus where the genetic material of the cells is packaged into chromo-
somes, which contain the blueprints for all of the proteins that make up the cell. These
proteins are manufactured outside the nucleus but within the cell body and have many
different roles. They may be used to build structures within the cell, be involved in
communication between or within cells, or control the manufacture of other molecules
required by the cell. The short branches that emerge from the cell body are called
dendrites and the single long process, the axon (fig. 1-3B). The axon carries informa-
tion in the form of pulses of electricity to other nerve cells, either nearby or at distant
sites within the nervous system or outside of it. Usually the axons from nerve cells
in the same region run together in bundles, and these form the white matter. This can
therefore be considered the cabling of the nervous system. The axons finally end in a
10 Chapter One

series of fine branches, which contact the dendrites and cell bodies of other neurons.
Information is passed from the axon to the dendrite by the release of chemicals called
neurotransmitters. Generally these will either excite or inhibit the follower cell. The
dendrites therefore represent the input end of the neuron, and the axon, the output
end. The contacts between neurons are found in the gray matter and consequently,
this is where the nervous system carries out all of its complex functions. In the spinal
cord, the gray matter makes up the central core while the white matter runs along the
periphery; however, in many regions of the brain it is the other way around.

Nerve Cells in the Spinal Cord

Nerve cells can be divided into three broad categories: motor neurons, sensory neu-
rons, and interneurons. We will examine each of these in turn.
Motor neurons. These have their cell bodies in the ventral region (ventral horn)
of the gray matter within the spinal cord, which faces the front when we are standing
upright. The axons of the motor neurons leave the spinal cord through the ventral roots
and pass into the segmental nerves, which convey them to the muscles. The axons of
motor neurons can be very long. A motor neuron supplying a muscle in the hand will
have its cell body in the cervical (neck) region of the spinal cord, so its axon will be
more than a meter long. As all of the proteins needed by the cell are manufactured in
the cell body, they must be transported along the entire length of the axon if they are
required at its terminal end. The cell must therefore not only construct and maintain
the axon but also maintain an active transport system, which requires energy to run it.
To put this task into perspective, if we imagine the motor neuron cell body to be the
size of a tennis ball, the axon running to a muscle in the hand would at this scale be
about two kilometers long!
Many motor neurons are needed to control each muscle and together these are
known as its motor pool. Every motor neuron contacts many muscle fibers. The group
of fibers driven by one motor neuron is called a motor unit (fig. 1-4); however, these
are nonoverlapping as each muscle fiber is driven by only a single motor neuron. The
motor neurons within the motor pool are not equivalent. Some control large motor
units composed of thousands of fibers, and others much smaller ones. When fine
movements requiring little strength are needed, the small motor units are activated
first. The strength of contraction can be gradually increased by recruiting additional
small units. As the contraction increases further, the increments of force gradually
get bigger because the newly recruited motor units become increasingly large. This is
called the size principle of motor recruitment and makes the most efficient use of the
motor neurons in the pool, as the greatest degree of control is needed when the forces
generated by the muscle are small. In a strongly contracted muscle, adding a small
motor unit would have little effect and so it is most appropriate to recruit large motor
units under these circumstances. The number of motor neurons needed in the motor
pool of a particular muscle depends on what it is used for, not simply on its absolute
size. For example, a small lumbrical muscle in the hand that contributes to its dexterity
and needs fine control has about a hundred motor neurons, whereas a muscle in the
shin, though two hundred times as powerful, has only seven times as many (Hamilton,
Jones, and Wolpert 2004). A single one of the six muscles that move the eye, which
 Introduction to the Tissues of the Body 11

Figure 1-4. The control of muscles by motor neurons. Muscles are composed of large
numbers of long, thin muscle fibers that contract when they are stimulated by motor neu-
rons. Each muscle is supplied by tens of thousands of motor neurons that constitute its mo-
tor pool. Each motor neuron contacts a small number of muscle fibers that make up what
is called a motor unit. Because every muscle fiber is supplied by only one motor neuron,
the motor units are controlled entirely independently.

must be controlled very precisely, has nearly three thousand motor neurons (Feinstein
et al. 1955). The size principle has considerable significance for instrumentalists
because it means that the greatest control over movement is achieved only when the
forces used are at a minimum. This is just one of several reasons why a conscious ef-
fort should be made to relax the fingers as much as possible during playing in order
to eliminate excessive or inappropriate muscle contraction.
Sensory neurons. The second class of nerve cell that we need to consider is the
sensory neuron. These neurons carry information from the skin, muscles, joints, and
organs to the central nervous system. They have an unusual structure, however, as
the cell body does not lie at the end of the axon, but on a side branch part of the way
12 Chapter One

along its length (fig. 1-3). The cell bodies form a cluster known as the dorsal root or
segmental ganglion. The axons then continue on their way to the spinal cord, where
they end mostly in the dorsal region of the gray matter (the dorsal horn).
There are several categories of sensory neurons, each of which carries a particular
type of sensation. We will first consider the nerve cells that supply the skin and are
concerned with temperature and pain and with different types of tactile sensation. The
information they carry is described as exteroceptive because in the main it concerns
the interaction of the body with the external world. Figure 1-2B is a diagram of the
skin showing the peripheral endings of the main classes of exteroceptive sensory neu-
rons. Some of these axons have unadorned, naked endings and others are encapsulated
by other types of cells that modify their response properties. Some naked endings
respond to painful stimuli or to temperature, while others wrap around hair follicles
and are activated by hair movement. In the superficial layers of the skin, two types
of encapsulated endings are found, Meissner’s corpuscles and Merkel cell endings.
The sensory nerve cells associated with Meissner’s corpuscles respond transiently to
indentation of the skin and so can detect brief, rapidly changing contact. Merkel cell
endings have a more prolonged response to contact. Because the endings of these
nerve cells are clustered close together in the most sensitive regions of the body, such
as the fingertips and the lips, they provide detailed information of the shape or tex-
ture of the surface that is making contact with the skin. They would therefore allow
the precise position of a string or a key under the pad of the finger to be monitored.
In the deeper layers of the skin there are two additional categories of encapsulated
sensory endings, the Pacinian and Ruffini corpuscles. Pacinian corpuscle endings are
sensitive to vibration such as that from a string or from the surface of the instrument.
Ruffini endings respond to shear stresses generated when the outer layers of the skin
are being dragged parallel to the surface, creating tension with the deeper layers. This
might occur when our grip is slipping or in string playing, when the finger is rocked
to create vibrato. The relevance of these different exteroceptive sensations for musi-
cians is therefore obvious.
Not all sensory nerve cells respond to exteroceptive information. An important
component of sensation originates from the body itself. This is known as proprio-
ception. Proprioceptors (fig. 1-3A) provide information about the degree of muscle
contraction and the level of tension in tendons (for example, as a consequence of the
weight being supported) and the angle at which a joint is being held. We do not need
to look at our arms or fingers to have an awareness of their position. Proprioception
plays a major role in such basic activities as the maintenance of balance as well as
in the control of highly skilled movements. Proprioceptive feedback is particularly
important when we are learning a new activity such as playing an instrument. For
many instruments, one of the crucial early skills that must be acquired is the ability to
position the hands and fingers accurately without looking at them, and this depends
on proprioception. It is also important when learning the fingering of a new piece of
music. After the brain has sent the signal to activate the muscles, their performance
is monitored by proprioception. Through repetition, the movements are refined by
comparing what is actually happening with what was intended. Though part of this
monitoring process comes from what we hear, a significant part also comes from how
 Introduction to the Tissues of the Body 13

the movement feels. An internal image of technique (motor memory) is thus gradu-
ally built up until the movements become accurate and automatic. This is particularly
crucial for violinists, cellists, and viola players, who must devote a significant propor-
tion of practice time to establishing the precise positioning of the fingers on which
accurate intonation depends.
Interneurons. The sensory neurons and motor neurons carry information to and
from the spinal cord, respectively, but by far the most abundant nerve cells within
the nervous system are those that connect neurons. These are called interneurons (fig.
1-3). Within the spinal cord they are necessary for many simple activities such as re-
flexes. They also carry information from the cord to the brain and vice versa. Though
we will not consider the details of how they operate, they are responsible for all the
complex functions that are carried out by the central nervous system, some of which
are discussed in chapters 8 and 9.

The Peripheral Nervous System

The peripheral nervous system is composed of the nerves that run from the spinal cord
to supply the rest of the body. These nerves therefore carry the axons of sensory and
motor neurons, but in addition they contain axons belonging to what is known as the
autonomic nervous system. This component of the nervous system will be dealt with
in chapter 10 as it plays an important role in the response of the body to stress. The
neurons of the autonomic nervous system control the smooth muscle of the organs and
arteries. They also supply the sweat glands of the skin, the secretory tissue that keeps
the lining of the nose and throat moist, the salivary glands, and glands in the stomach
and intestine that secrete digestive enzymes.
Bundles of axons are easily damaged and so nerves are structured to give them
protection. Where they pass in narrow channels close to the bone or between muscles,
they may be compressed or trapped when the muscles contract, and where they cross
joints they may rub against the bone or be stretched significantly as the joint is flexed
or extended. The major nerves therefore have a tough outer sheath that helps to limit
the degree of stretching, while within they are reinforced by fibers such as collagen.
To prevent the axon bundles from being crushed, the spaces between them are packed
with fatty tissue that plays a similar role to bubble wrap. Small nerves, such as those
running along the sides of the fingers, are less well protected and are sometimes
pinched against the bone when a string is plucked or by the pressure needed to support
an instrument. Unless severed completely, nerves that are damaged will often regrow,
but this is a slow process as it runs at a rate of about one millimeter per day.

THE NATURE OF PAIN AND ITS TREATMENT

Because pain is a major symptom of many performance-related physical problems,

it is important to understand the nature and properties of its various manifestations.
Sensations that we interpret as pain are carried by a particular class of sensory nerve
endings that require very strong stimuli to activate them, that is, they have a high
14 Chapter One

threshold. Biologists and clinicians make a clear distinction between nociception,

which is the perception of intense sensory stimulation indicative of tissue damage,
and pain itself, which is the unpleasant subjective sensation that this creates in our
minds. Whether we perceive a noxious stimulus as pain depends on our mental state.
A serious wound incurred in battle or when fleeing from danger may not feel pain-
ful until much later, when the immediate threat has diminished. On the other hand,
quite innocuous physical stimulation of the skin may under some circumstances feel
exceedingly painful. In the context of performance-related injury, pain may be made
worse by anxiety or by changes that occur in the central nervous system when a nox-
ious signal lasts for a prolonged period of time.
The electrical pulses that run along all sensory nerves are identical, regardless of
whether the sensation that gives rise to them is perceived as pleasant, unpleasant, or
neutral. If we use the analogy of a telephone network, each type of sensation (modal-
ity) has its own line carrying similar electrical signals that represent words, regardless
of their import. However, though the line that carries nociceptive information is simi-
lar to the rest, it is given a priority because it runs to the “red emergency telephone.”
This is the one we cannot afford to ignore. It always gets our attention and triggers an
alarm that the brain labels as pain. The sensation of pain therefore originates there and
not in the peripheral nervous system or spinal cord.
There are two phases to nociception. When we suffer an injury such as a cut or a
blow, there is an initial sharp, well-localized pain that comes from rapidly transmit-
ting sensory nerve fibers (Aδ fibers). This is followed by a duller and more diffuse
pain, carried by slower conducting sensory fibers (C fibers), that lasts for a longer
period of time. The C fibers also respond to noxious heat but can be stimulated by the
active chemical ingredient in chili powder (capsaicin), which interacts with the same
proteins in the nerve ending that are influenced by heat. In other words, to the nervous
system, real heat and the effect of capsaicin are identical. Capsaicin can be used to
deactivate these sensory fibers and this is why people who eat a lot of spicy food cease
to be strongly affected by it. In some situations capsaicin is used therapeutically to
reduce the sensitivity to pain.
Pain is essential to normal life, as it often prevents severe injury by alerting us to
dangerous situations; for example, it stops us from drinking very hot liquids, from
maintaining skin contact with sharp or hot objects, or from putting weight on injured
joints and ligaments. Low levels of pain or discomfort encourage us to change posi-
tion regularly, whether awake or asleep, thus preventing skin pressure sores or the
maintenance of stressful postures.
Nerve cells communicate with each other using chemicals called neurotransmitters.
In the case of sensory nerve cells, this is the amino acid glutamate. However, the sen-
sory neurons of the pain system also contain small proteins (peptides) such as substance
P or CGRP, which are released within the gray matter of the spinal cord by the axons of
sensory neurons, together with glutamate. Indeed, the sensation of moderate to severe
pain is dependent on this release of substance P. The peptides are also released from
the other end of the sensory neurons, at the endings that lie within the skin, muscles,
and joints. Here, substance P induces the release of other chemicals (e.g., histamine)
from cells within the damaged tissue, and together these not only promote inflamma-
 Introduction to the Tissues of the Body 15

tion but also increase the sensitivity of nociceptive sensory endings to further stimula-
tion. This is known as hyperalgesia. Some nociceptive nerve endings are completely
unresponsive to mechanical stimulation until they have been sensitized in this way. In
addition, other compounds promoting inflammation and hyperalgesia are formed from
the breakdown of damaged cells at the site of injury. Among these are the prostaglan-
dins. Painkillers such as aspirin and ibuprofen block the production of prostaglandins
and so not only reduce pain, but also limit inflammation. For this reason they are often
described as nonsteroidal anti-inflammatories (NSAIDs) to distinguish them from the
steroid compounds that are used to reduce inflammation alone.
The effect of prolonged stimulation of nociceptive sensory nerve cells is to cause
a buildup of substance P within the spinal cord. This leads to a progressive increase
(windup) in the activity of the pain pathways, which increases the sensitivity to pain
(secondary hyperalgesia). In addition, sensory endings from the undamaged area
around the wound, endings that normally carry innocuous sensations perceived as
touch, may start to generate the sensation of pain. This is known as allodynia and is
the result of changes in the circuitry within the spinal cord that handles sensation.
Normally this quickly returns to normal as the peripheral swelling resolves and the
injury heals; however, in long-term (chronic) pain states it can become very difficult
for the spinal circuits to switch back to more normal states of sensation.
Though, as described above, different types of sensation run to a large extent in dis-
tinct channels, there is nevertheless a degree of interaction between them. This is cen-
tral to the well-known “gate theory” of pain processing. This proposes a mechanism
by which the access of the nociceptive sensory signals to the pain pathways that ascend
from the spinal cord up into the brain can be blocked under certain circumstances. One
way in which this can be achieved is by the simultaneous intervention of a mild sen-
sory stimulus. We are all familiar with this process. When we collide with some hard
object, we rub the affected area to close the “gate” and hence lessen the pain. The gate
can also be closed by the action of drugs such as opium and its derivatives (e.g., lauda-
num, morphine), which have been used for centuries to control severe pain. How this
is achieved has only quite recently been discovered. It transpires that the nervous sys-
tem contains its own opium-like chemicals called endorphins and enkephalins. These
are neurotransmitters used by nerve cells that block pain by preventing the transfer
of information from the nociceptors to the spinal pain pathways. Morphinelike drugs
mimic this action but in a more powerful way. The ability of our mental state to influ-
ence pain perception comes about partly via pathways that descend from the brain to
stimulate the endorphin- and enkephalin-containing nerve cells. Descending pathways
also use other means to close the pain gate, such as reducing the buildup of substance
P in the spinal cord. However, the brain can not only reduce our sensitivity to pain, it
can also enhance it. If you are apprehensive about receiving an injection, it will be a
more painful experience than if you are unconcerned, or if the nurse skillfully distracts
your attention at the moment the needle goes in. Pain perception is not only influenced
by expectation but also by mood. For example, anxiety and depression both enhance
pain perception, as does a negative outlook and tendency to expect the worst (some-
times known as catastrophization). People with a more positive outlook, and greater
ability to cope with life’s problems, are more likely to have higher thresholds to pain.
16 Chapter One

This may be linked to a feeling of having a degree of control over the situation that
leads to the pain. For this reason, psychological as well as pharmacological strategies
have an important role in the treatment of chronic pain states.
A comprehensive account of the current understanding of the phenomenon of pain
can be found in the definitive book on the subject by Melzac and Wall (Wall, Mc-
Mahon, and Koltzenburg 2006). Though this weighty tome contains a great deal of
information about the origin and treatment of different types of pain, it also makes
clear just how much we do not yet understand about pain states and how to manipulate
them.

REFERENCES

Blum, J., and G. Ritter. 1990. Violinists and violists with masses under the left side of the angle
of the jaw known as “fiddler’s neck.” Med Prob Perform Artists 5 (4):155–60.
Feinstein, B., B. Lindegard, E. Nyman, and G. Wohlfart. 1955. Morphologic studies of motor
units in normal human muscles. Acta Anat (Basel) 23 (2):127–42.
Gambichler, T., S. Boms, and M. Freitag. 2004. Contact dermatitis and other skin conditions
in instrumental musicians. BMC Dermatol 4 (1):3.
Hamilton, A. F., K. E. Jones, and D. M. Wolpert. 2004. The scaling of motor noise with muscle
strength and motor unit number in humans. Exp Brain Res 157 (4):417–30.
Harvell, J., and H. I. Maibach. 1992. Skin diseases among musicians. Med Prob Perform Art-
ists 7:114–20.
Okner, M. A. O., T. Kernozek, and M. G. Wade. 1997. Chin rest pressure in violin players:
Musical repertoire, chin rests and shoulder pads as possible mediators. Med Prob Perform
Artists 12:112–21.
Onder, M., A. B. Aksakal, M. O. Oztas, and M. A. Gurer. 1999. Skin problems of musicians.
Int J Dermatol 38 (3):192–95.
Thomas, P., F. Rueff, and B. Przybilla. 2000. Cheilitis due to nickel contact allergy in a trumpet
player. Contact Dermatitis 42 (6):351–52.
Wall, P. D., S. B. McMahon, and M. Koltzenburg. 2006. Wall and Melzack’s textbook of pain.
5th ed. Philadelphia: Elsevier/Churchill Livingstone.
 Chapter Two

Posture and the Back in

Musical Performance

Even in players with good posture, many instruments impose asymmetrical stresses on
the muscles of the back, and at professional as well as amateur levels, many perform-
ers make life difficult for themselves by assuming unnecessarily tortuous positions
while playing. If you watch the posture of musicians when they return to the stage
to acknowledge the applause, you will often see that they still stand asymmetrically
when not playing (or even carrying) their instruments. In order to avoid playing-
related injury it is therefore important not only to minimize the stresses on the body
while performing, but also to reduce or eliminate the spillover of playing posture into
everyday life. It can become a rather tedious habit to find yourself being distracted
from the music by analyzing the postures of performers on stage, particularly if you
insist on sharing your observations with your companions; however, it is very instruc-
tive. Large ensembles such as orchestras are the most interesting in this regard, as
they allow a direct comparison between players of the same instruments. It will soon
become obvious that postures vary widely and some are inherently more risky than
others in terms of potential injury. When you have read the material in this chapter
and the next, much of this will become clear to you, as the principles behind cor-
rect posture are quite straightforward. Promoting safe and sustainable posture is also
fundamental to optimizing musical performance and should be a component in instru-
mental teaching from the earliest stages. This is much easier to instill right from the
beginning than to correct later. In this chapter we shall start by exploring the anatomy
of the back. We will then look in some detail at the origins of back pain, how this is
related to the playing posture of different instruments, and how it can best be avoided.
Finally, we shall examine the vexing question of seating, a subject that is only just
beginning to get the attention it deserves.

STRUCTURE OF THE VERTEBRAL COLUMN (SPINE)

The spine or vertebral column is composed of a series of thirty-three bones (vertebrae)

that run from the base of the skull to the pelvis (fig. 2-1A). These are divided into five

17
18 Chapter Two

Figure 2-1. The vertebral column and pelvis as seen from the side. A. There are seven cervical
vertebrae in the neck, twelve thoracic vertebrae each supporting a pair of ribs, five lumbar verte-
brae, five sacral vertebrae that are fused to the pelvis, and four coccygeal vertebrae that form a
single bony mass. Each section of the vertebral column is curved in a characteristic fashion when
the body is held erect. B. The root of a large segmental nerve emerges from between each pair of
vertebrae. Many of the cervical roots merge and branch in a complex fashion to supply the arm;
an equally complex system of lumbar and sacral roots supplies the leg. C. A more detailed view
of the origin of the segmental nerves as they emerge from the central space within the vertebral
column, which carries the spinal cord.

groups. The neck is supported by the upper seven or cervical vertebrae; the twelve
thoracic vertebrae are each attached to a pair of ribs; five lumbar vertebrae support the
abdomen; the five sacral vertebrae are immobile, being rigidly fixed to the pelvis; and
the final four vertebrae are fused together to form the coccyx, a rudimentary tail-like
structure. Figures 2-1C and 2-2A, B, show some of the typical features of the verte-
brae. The central element is a short column of spongy bone called the vertebral body,
which carries the weight of the trunk. The area of the vertebral body increases progres-
sively from the first cervical to the last lumbar vertebra as the length of the column
and the weight that it must bear increases. Behind the vertebral body is a channel that
contains the spinal cord and this is protected by an arch of bone (the neural arch). A
pair of segmental nerves arise from the spinal cord at the level of each vertebra and
emerge through openings on each side of the neural arch (figs. 2-1B, C). A number of
bony projections arise from the arch. These provide attachment sites for ligaments and
muscles that support the spine. They also carry the smooth articular surfaces of some
of the joints that link adjacent vertebrae or support the ribs. Usually there is one dorsal
process called a spine, which sticks out from the apex of the arch (and therefore points
toward the skin surface), and one transverse process emerging from each side.
The flat upper and lower surfaces of the bodies of adjacent vertebrae do not make
direct contact with one another. Instead they are separated by the cushionlike interver-
tebral disks, which act as shock absorbers (fig. 2-2). These have a tough, fibrous, ring-
like outer wall (the annulus fibrosus) composed of several layers of cartilage whose
fibers run in different directions like the reinforcement of a cross-ply tire. The annulus
fibrosus is firmly attached to the vertebral bodies above and below. At the center of
Figure 2-2. The lumbar vertebrae and their disks. A, B. One of the lumbar vertebrae viewed (A)
from the side and (B) from above. The main load-bearing parts of the vertebrae are the vertebral
bodies. These sit one above the other, separated by the intervertebral disks. The disks have a
distortable, gelatinous core (nucleus pulposus) contained by a fibrous wall (annulus fibrosus)
that acts rather like the wall of a tire. Behind this lies the channel containing the spinal cord.
The spine and transverse processes provide attachment sites for the muscles and ligaments of the
back. Adjacent vertebrae are linked by several small joints that help maintain alignment when
the vertebral column is flexed, extended, or rotated. C–G. Views of the upper lumbar vertebrae
as seen from the side (C–E) and directly in front (F–G) showing how the nucleus pulposus of an
intervertebral disk is distorted as the vertebrae move relative to one another. H. If the annulus
ruptures, the nucleus pulposus may be extruded so that it compresses either the spinal cord or
the nerve roots emerging from it. This is known as a prolapsed or “slipped” disk.
20 Chapter Two

Figure 2-3

the disk is a deformable gel-filled core called the nucleus pulposus. In a young adult,
the nucleus pulposus is 90 percent water with the remaining 10 percent being made
up of water-absorbing proteins. Water is incompressible, and so provides a good
buffer between the bony surfaces of the vertebral bodies, but the wall of the nucleus
pulposus is flexible and can be distorted as the angle between the vertebrae changes
when the spine is tilted forward or back, or from side to side. The degree of distortion
that is possible in the nucleus pulposus is limited by the annulus fibrosus. In effect,
 Posture and the Back in Musical Performance 21

Figure 2-3. A–C. Ranges of spinal movement. The movements of the vertebral column can be
described as flexion, extension, rotation, and lateral bending (A, B). The total range of movement
is the sum of the movement possible at the joints between adjacent vertebrae, shown graphically
in C. The joints of the thoracic vertebrae (T1–T12) allow the least movement, partly because of
the rigidity of the rib cage, while those of the neck (C1–C7) have the most freedom. The joint
between the first two neck vertebrae (C1–C2) allows the greatest degree of rotation (about 40° to
either side). The lumbar vertebrae bear the weight of the trunk but are supported only by muscles,
and so although considerable flexion and extension and a smaller amount of lateral bending are
possible, rotation is severely limited by the shape of the small joints between them, thus impart-
ing some degree of stability. D–G. Standing posture. In the diagrams shown here and in figure
2-5, the dashed reference lines centered on the hip joint indicate the degree of pelvic rotation in
various postures relative to the joint’s position in an erect stance. D. When standing fully erect,
maximum height is attained. The center of gravity of the trunk lies in front of the vertebral column
and so the trunk must be held erect by the active involvement of muscles of the back, abdomen,
and legs (arrows). If the muscles are allowed to relax, the abdominal wall becomes more promi-
nent and the pelvis rotates forward. This increases the curvature of the lumbar spine (increased
lordosis) and the shoulders move forward, resulting in a reduction of height. F. When the trunk
bends forward during standing, the first 60° of the movement is brought about by flexion of the
lumbar spine, which reverses its curvature (i.e., it becomes kyphotic). G. In normal individuals,
a further 25° of flexion can be achieved by allowing the pelvis to rotate forward. This rotation is
limited by the length of the hamstring muscles, which run from the back of the pelvis to below
the knee. Further rotation requires flexing of the knee to give the muscle more play.

the vertebrae roll over the nucleus pulposus as if it were a rubber ball sitting at the
center of the disk, giving the spine its flexibility. The compression acting on the disk
when we are upright forces some of the water out, and though the nucleus can counter
this to some degree by absorbing salts to draw water in again by osmosis, the spine
typically shortens by a couple of centimeters in the course of a normal day. When we
lie down at night, the pressure on the disk is relieved and the water content increases
again. As we get older the proportion of water in the nucleus pulposus is reduced by
approximately one-third and this, together with the fact that older people tend to stand
less erect, is one of the main reasons height declines with age.
If the vertebrae were only linked via the intervertebral disks, the spine would be
very unstable, and constant and considerable muscular effort would be required to
hold it erect. In practice, the degree to which adjacent vertebrae can move relative to
one another is restricted by the presence of smaller accessory joints (zygapophysial
joints) that link vertical flanges that rise from the neural arches of adjacent vertebrae.
These joints are stabilized by short ligaments that limit their range of movement and
further reduce the amount of muscular work needed to maintain an erect posture.
The degree of movement that is possible between adjacent vertebrae varies consid-
erably in different regions of the spine. Bowing forward is described as flexing the
spine, while bending backward is known as extension (fig. 2-3A). The first cervical
vertebra (the atlas) lacks a vertebral body, but the shape of the load-bearing joints that
support the skull allows significant flexion and extension movements (as in nodding)
but virtually no rotation of the head on the neck (turning the head from side to side).
In contrast, the joint between the atlas and second cervical vertebra (the axis) is the
one at which the most rotation can take place. This allows the head to turn through an
angle of about 40° to each side (fig. 2-3C). The thoracic vertebrae carry the ribs; at the
22 Chapter Two

front of the chest these are linked either to each other or to a flat plate of bone called
the sternum (breastbone) (fig. 2-4A). The linkages between the ribs and the sternum
are made of flexible cartilage, but the box formed by the chest wall is quite rigid. This
considerably restricts the range of movement possible between the thoracic vertebrae
(fig. 2-3C). It might be expected that the lumbar vertebrae would have a much greater
freedom of movement, and this is certainly true for flexion and extension; however,
the amount of side-to-side bending and the degree of rotation possible at each joint are
actually rather small. This is a consequence of the position and orientation of the fac-
ets of the accessory joints between the vertebrae. If the degree of movement between
the lumbar vertebrae were greater, it would put intolerable demands on the muscles
of the back that maintain lumbar stability. The total flexibility of the spine in any one
direction is of course the sum of the movement possible at each joint, so even though
the relative motion between pairs of adjacent vertebrae is small, when this is summed
over the twenty-five movable joints that lie between them, the total angle through
which the spine can be bent is considerable. The head can be flexed and extended
on the cervical vertebrae through a total angle of 100°, while for the entire vertebral
column this angle can be as much as 250° (Kapandji 2000), though we generally see
such extreme bending only in gymnasts.
Even in a person with normal posture who is standing or sitting erect, the spine
is far from straight. If the body is held in what is known as vertical alignment (with
the shoulder joint lying on a vertical line above the hip and ankle joints), the spine
exhibits several pronounced curvatures. These are described by the terms kyphosis
and lordosis. If we think of the curves as being shaped like the letter “C,” kyphosis
refers to the situation where the opening of the C faces the front of the body, while
in lordosis the opening faces the back (fig. 2-1A). When we stand erect, the cervical
spine has a slight lordosis, the thoracic spine has a clear kyphosis, the lumbar spine
exhibits a marked lordosis, and the fused sacral and coccygeal vertebrae are together
kyphotic. This can change when other postures are adopted. When we sit in a chair
and slump forward, the lumbar spine straightens and ultimately becomes kyphotic.
Somewhat confusingly, while many writers apply the terms kyphosis and lordosis to
the normal curvatures of the erect spine as I have done here, others use them only to
describe curvatures that deviate from the norm. When seen from directly behind, the
spine should appear straight. A strong lateral curvature of the spine (scoliosis) is usu-
ally symptomatic of disease, though small lateral deviations can develop as a result of
everyday activities such as habitually carrying a heavy bag by a strap over one shoul-
der. If this posture is maintained, it can lead to discomfort or pain. We will return to
this topic later when discussing the playing postures of instrumentalists.
The vertebrae are held in position by a series of ligaments that help to limit the
potential displacement between them. Some of these run along the walls of the cen-
tral space that carries the spinal cord. One such ligament (the ligamentum flavum)
runs down the roof of the neural arches beneath the spines of the vertebrae. Unlike
most of the other ligaments, which are relatively inextensible, the ligamentum fla-
vum is elastic and thus holds the spine under tension when it is in the erect position.
A long, broad external ligament links the front of the bodies of the vertebrae for the
entire length of the vertebral column, while a narrower one links the back of the ver-
 Posture and the Back in Musical Performance 23

tebral bodies. Short external ligaments run between adjacent spines and transverse
processes. The cervical vertebrae receive additional support from a large sail-like
ligament (the nuchal ligament) that passes downward from the back of the skull.
This not only braces the vertebrae of the neck, but also provides an attachment site
for several muscles.
We have already seen that the annulus fibrosus of the intervertebral disks links the
bodies of vertebrae. The smaller joints that link the vertebrae to each other and to
their ribs are held together by sheets of tissue that form a joint capsule (see chapter
1). These capsules not only limit the vertebrae’s range of movement but also serve to
contain the synovial fluid, which lubricates the joint. Though the spinal ligaments are
important in keeping the vertebrae in alignment, they do not impart much strength to
the vertebral column which, being composed of a series of blocks stacked one above
the other, remains quite unstable. The load that would collapse the unsupported spine
is several hundred times less than the weight it must support as we carry out our nor-
mal daily activities; it is only through the action of muscles that the necessary stability
can be achieved. In addition, even when we stand erect, the center of gravity of the
trunk lies in front of the vertebral column, which must be actively braced by muscles
against a force that is trying to flex it forward.

Figure 2-4. The basic organization of the postural muscles of the back. A. The vertebral col-
umn seen from the front showing some of the major muscles acting on the head and neck. The
longus capitus muscles act on the head, pulling it downward, while the longus colli flex the neck
forward. The sternocleidomastoid, which rotates the head to the opposite side and upward, runs
from the skull, just behind the ear, to the collarbone and sternum. The psoas muscle runs from
the side of the lumbar vertebral bodies to the femur and can act to increase lumbar lordosis. The
quadratus lumborum runs from the transverse processes of the lumbar vertebrae to the edge of
the pelvis and can flex the trunk to the side. B. The vertebral column from the back. The deep
muscles of the vertebral column are collectively called transversospinalis. They stabilize the indi-
vidual vertebrae and contribute to their ability to move relative to one another. C. The superficial
muscles that act directly on the vertebral column are collectively known as the erector spinae.
These, and the splenius capitus muscle, are the major muscles responsible for extending the spine
when active on both sides and for rotating the spine when acting on one side only.
24 Chapter Two

MUSCLES ACTING ON THE SPINE

The basic layout of the major muscles acting on the vertebral column is shown in
figure 2-4.

The Neck
We will first consider the muscles of the neck, which can be divided into three groups:
1) those that link the vertebrae to each other, 2) those that run between the vertebral
column and the base of the skull, and 3) those that are not attached to the vertebrae
but act on the neck indirectly.
Attached to the front of the neck vertebrae are the prevertebral muscles (fig. 2-4A),
which flex the neck (i.e., pull the head forward and down). The longus colli series
(literally, the long muscles of the neck) run between the cervical and upper thoracic
vertebrae and therefore act only on the neck, while the longus capitus muscles (the long
muscles of the head) are attached at one end to the cervical vertebrae and at the other
to the base of the skull and therefore act mainly to tilt the head forward. In addition,
two small rectus (straight) muscles run between the skull and the first of the cervical
vertebrae to stabilize the head. The head and neck are also moved by two large muscles
that run like guy ropes from behind the ear to the collarbone and the sternum. These are
the sternocleidomastoid muscles, which can often be seen forming a V-shaped ridge
in people with thin necks. Because their attachment to the skull lies behind the axis of
rotation and the point of balance of the head on the neck, they turn the face upward
when they contract together. If only one is active, the face is also turned to the opposite
side. Because of their attachment to the sternum, the sternocleidomastoids can also be
considered accessory muscles of respiration (see chapter 4).
The organization of the muscles at the back of the neck, which extend it (tilting
the head upward), is rather more complicated. One set of muscles (e.g., the upper
muscles of the transversospinalis and erector spinae groups) runs from the cervical to
the upper thoracic vertebrae and act only on the neck, while another set (e.g., splenius
capitus) runs from the vertebrae to the back of the skull and tilts the head backward
(fig. 2-4B, C). In addition, there are two sets of small rectus and oblique muscles that
run from the upper two cervical vertebrae to the back of the skull to stabilize the head.
A number of muscles also connect the vertebral column to the back of the skull and
the shoulder blade (scapula). The shoulder blade is not directly attached to the trunk
but is suspended from it by several sets of muscles and is therefore highly mobile
(see chapter 3). The position of the shoulder blade has considerable significance for
the playing posture adopted by violin and viola players, and so some of the muscles
attached to it will be considered here (see fig. 3-3). Of particular significance in this
context are those that raise the left shoulder and pull it inward when the instrument
is held under the chin. The shoulder blade is shaped like an inverted triangle, with a
prominent ridge or spine running approximately horizontally across the back. Sets of
small muscles, the levator scapulae, run down from the upper cervical vertebrae to
the inner corner of the triangle and pull it upward. Below these are two broad rhom-
boid muscles that pull the inner edge of the shoulder blade toward the spine. Over
 Posture and the Back in Musical Performance 25

the surface of these lies a triangular sheet of muscle called the trapezius, the upper
part of which runs from the back of the skull and the nuchal ligament to the spine of
the scapula and is active when the shoulders are shrugged. As these shoulder muscles
are attached at each end to a movable structure (the scapula or the vertebral column),
their effect depends on which end is held in a fixed position. Therefore, although
they are usually described in terms of their effect on the scapula, if this is held fixed
by the action of other muscles, it is the head or neck that is moved. This is an impor-
tant general principle of muscle action that is sometimes not made clear in standard
descriptions of their actions, and we will encounter it again later. In fact, in many
actions, there is a degree of movement of the bones at both ends of such muscles,
within the tolerances set by the actions of other muscles, ligaments, and joints. From
the foregoing account it will be appreciated that there are many superimposed layers
of muscles in the neck and shoulders. These are often chronically and unnecessarily
held in tension as a result of emotional stress, and many tension-relieving strategies
(e.g., massage) pay particular attention to this area.

The Thoracic and Lumbar Spine

There are no muscles equivalent to longus colli running on the front surface of the
vertebral bodies of the lower thoracic or lumbar vertebrae. This region of the vertebral
column is flexed mainly by the muscle that forms the front wall of the abdomen, the
rectus abdominus (popularly known as the “six pack”), which runs between the lower
ribs and the midregion of the pelvis (figs. 2-3D, 4-3). In this it is aided by the action
of the psoas muscle (fig. 2-4A), which runs from the lateral surfaces of the bodies of
the lumbar vertebrae to the top of the femur (thighbone). The psoas can also move
the thigh, but when the femur is carrying the weight of the body, its position is fixed,
so the action of the psoas is to flex the lumbar spine. On either side of the rectus ab-
dominus, the muscular wall of the abdomen is formed by three sheets of oblique and
transverse muscle fibers (fig. 4-3). The abdomen is a sealed chamber, so if all of the
muscles in its walls contract at the same time, the pressure within it rises, and as its
circumference is reduced, the visceral organs are pushed up toward the chest cavity. If
the breath is held at the same time, keeping the volume of the chest constant, the trunk
becomes rigid, like an inflated ball, and can provide support to resist forward flexion
of the spine. This is a common strategy we use when trying to lift a heavy weight from
the floor and is seen in its most extreme form in weightlifters.
The arrangement of the muscles on the dorsal surface of the back is quite compli-
cated; however, they are usually divided into a deep group and a superficial group (fig.
2-4B, C). The muscles of the deep layer run between the transverse process of one
vertebra and the spinous or transverse process of a vertebra several segments above.
These muscles stabilize and extend the vertebral column when active on both sides at
the same time and cause rotation when active on one side only. The more superficial
muscles, collectively called the erector spinae, run between the spinal processes of the
vertebrae, or between the bases of the ribs. When the erector spinae on both sides are
active simultaneously, they cause extension of the vertebral column, and when active
on one side only, lateral bending. When the actively extended back is viewed from
26 Chapter Two

behind, two muscular columns are visible beneath the skin on either side of a central
groove. The groove represents the position of the spinous processes of the vertebrae
and the columns, the collective bulk of the deep and superficial muscle masses. In
addition to these muscles, a broad sheet (the quadratus lumborum) runs from the back
of the pelvis to the transverse processes of the lumbar vertebrae and the base of the
lowest rib, forming part of the muscular wall at the back of the abdomen. Its action
contributes to the lateral bending of the lumbar spine.
If we bend forward from the waist, starting from the fully erect standing position
with the knees straight, the movement through the first 60° is that allowed by the
joints between the vertebrae. The weight of the trunk is supported by progressively
increasing activity in the erector spinae and deep muscles of the back. The pelvis is
held fixed by the action of the large muscles of the buttocks (the gluteus maximus) and
the hamstring muscles at the back of the thighs (fig. 2-3D–G). The gluteus maximus
runs from the back of the pelvis to the femur while the hamstrings run from the back
of the pelvis to the lower leg, just below the knee. We normally think of these muscles
in terms of their ability to move the femur or flex the knee joint; however, when the
leg is fixed and the knee locked to carry our weight, these muscles instead rotate the
pelvis backward. When we need to bend the trunk forward beyond 60°, the gluteus
maximus and hamstring muscles relax, allowing the pelvis to rotate forward relative
to the femur by about 25°, the weight being borne by the elasticity of the muscles and
their tendons. The length of the hamstrings determines the maximum possible angle
of forward pelvic rotation. If the knees are bent, this slackens the hamstrings, allowing
the pelvis to be rotated much further. For the same reason, the angle of the knees is
one factor that determines the pelvic rotation possible when we are sitting.
Because the center of gravity of the trunk lies in front of the vertebral column, there
is a tendency for it to flex forward during standing (fig. 2-3E). If this is allowed to
happen, the shoulders slump, causing the curvature of the thoracic spine to increase
(kyphosis). The pelvis rotates forward, increasing the lumbar curvature of the spine
(lordosis), and as a result, the abdomen bulges outward. When we stand fully erect
(fig. 2-3D), our height increases because the tendency to flex forward is resisted by the
muscles of the back. The rectus abdominus muscle helps to rotate the pelvis upward
under these circumstances, reducing lordosis. In young people, who generally have
good muscle tone, these corrective actions are easily maintained, but in older people
the loss of muscle strength makes this more difficult, particularly if the mass of the ab-
domen is increased by obesity. The slumped posture may, as a result, become perma-
nent, or at least habitual. This can be exacerbated by conditions such as osteoporosis,
which involves a general loss of bone mass throughout the skeleton and is particularly
significant for posture through its effects on the vertebrae. Osteoporosis is common in
older women, in whom it is often a consequence of the reduction of circulating steroid
hormones after menopause.
The majority of musicians do much of their professional playing while seated. The
posture of the spine in sitting is quite different from that in standing. This can be seen
in animation 2-1. An excellent review of the effect of sitting on the spine and its con-
sequences for seat design can be found in Harrison et al. (1999). Sitting in any posture
puts greater pressure on the intervertebral disks than does standing. This leads to a
 Posture and the Back in Musical Performance 27

greater loss of fluid from the nucleus pulposus. The main structures through which the
weight of the body is transmitted to the seat of the chair are the lowest points on the
pelvis. These are called the ischial tuberosities. The forces acting on these points while
we are seated erect on a hard chair rotate the pelvis backward, which reduces lumbar
lordosis (fig. 2-5A, animation 2-1). The use of seats with a lumbar support limits this,
reducing disk pressure and consequently the severity of the symptoms of low back
pain. If the body is allowed to recline farther backward from the erect position, the
lumbar spine will reverse its curvature to become kyphotic (fig. 2-5C, animation 2-1).
As a result, lumbar support becomes even more important for comfortable sitting. The
angle between the trunk and the thighs also has an effect on lumbar curvature through
the action of the postural muscles acting on the legs. It has been suggested that when
sitting erect, an angle of about 135° between the thighs and the trunk minimizes the
tension on the lumbar spine because it allows a reduction in the activity of the thigh
muscles, which act on the pelvis and spine (Keegan 1953), and increases lordosis.
This is possible because such a posture brings the center of gravity of the upper body
directly over the ischial tuberosities, minimizing the tendency of the weight of the
trunk to rotate the pelvis (fig. 2-5B, animation 2-1). Under these circumstances, the
degree of lumbar curvature lies between that of standing or sitting and approximates
to the degree of lordosis seen in the unloaded spine—the so-called neutral position
(Adams et al. 2002). To achieve such a posture, either the seat of the chair needs to
slope downward at an angle of 45° from horizontal, which would be rather impracti-
cal, or, alternatively, the sitter needs to move to its front edge to allow the thighs to
slope downward. Office chairs that allow the seat to be tilted typically have a range
of adjustment of 7°–15°.

BACK PAIN

About three-quarters of us experience back pain (particularly low back pain) at some
stage in our lives. This may reflect an obvious physical injury such as a muscle strain
or damaged disk, but often it is impossible to determine the precise source of the pain.
In some cases, what have been termed psychosocial factors appear to play a signifi-
cant role in back pain. For example, individuals who find themselves in unrewarding
situations (such as unstimulating jobs with poor working conditions) or have illnesses
such as depression are more likely to complain of physical problems, as the lack of
other compensations makes them more aware of their discomfort. In addition, one
symptom of stress is a general increase of muscle tension, particularly in the back,
neck, face, and jaw. Directly or indirectly, this may lead to postural problems and
back pain.
At one time it was widely accepted that an exaggerated lordosis would inevitably
lead to back pain, but objective studies have found little evidence for this (White and
Panjabi 1990). Apart from people employed in heavy manual labor, one of the groups
in whom back pain is most common is pregnant women. Here the pain derives from
the strain of supporting the weight of the growing fetus, which causes the center of
gravity of the abdomen to move farther forward from the vertebral axis. This may be
28 Chapter Two

exacerbated by the softening of ligaments, including those of the spine, that occurs
during pregnancy. In young musicians, occupation-related back pain is most likely to
be due to muscle or ligament strains related to maintaining playing postures for long
periods of time. In the majority of cases, low back pain is self-limiting, that is, with
rest from the stressful position it resolves itself over a period of time. Apart from
pain relief, the most suitable treatment is exercise aimed at strengthening the postural
muscles that support the back, with the aim of improving the alignment of the verte-
bral column and reducing the pressure on the disks (Horvath 2002; Paull and Harrison
1997; Richardson and Jull 1995). Bed rest, which was once a commonly prescribed
treatment, is now regarded as quite unhelpful as it causes rapid weakening of precisely
the sets of muscles that support the spine. Of course, rather than waiting for problems
to develop, it is better to adopt a posture that minimizes the risk of developing back
pain and to keep the postural muscles in good condition (see below).
Another type of soft-tissue injury that results in back pain is what is commonly
known as a “slipped” disk. This is an inappropriate term as the disk does not slip out
of position; instead, weakness or damage to the annulus fibrosus allows the nucleus
pulposus to push outward through it (fig. 2-2H). This most commonly occurs on
straightening the spine following a prolonged period of forward bending (flexion).
The disks that are most vulnerable are those of the lower cervical or lower lumbar ver-
tebrae. It is not unusual for more than one disk to be affected; the persistence of pain
following treatment is sometimes due to the fact that this has not been recognized. The
extruded nucleus pulposus almost always emerges from the back of the disk, and the
pain is the result of the pressure this exerts either on the spinal cord or on the nerve
roots that emerge from it. The nerve supply to the disks themselves is confined to the
outer part of the annulus fibrosus so that we normally have no sensation of pressure in
the core. However if the disk degenerates, the sensory nerve endings may invade the
nucleus pulposus, causing chronic pain (Freemont et al. 1997)

AVOIDING OCCUPATIONAL BACK PAIN

Lifting and Carrying

One of the reasons the muscles and ligaments of the back are vulnerable to injury is
that they run very close to the central axis of the vertebral column, which means that
they have a very poor mechanical advantage (i.e., they have a small lever arm) when
they act to resist flexion of the spine. As a result, the stresses they are subjected to
if the back is bent forward during lifting are extremely large. The farther away from
the body the weight being lifted is held, the greater the effective lever arm the weight
exerts on the spine and the greater the force that must be resisted by these muscles and
ligaments and the disks of the vertebrae. Vertical lifting should therefore be carried
out with a straight back using the large muscles of the thighs to provide the power
for the lift (fig. 2-5D). The weight should be held as close to the body as possible to
reduce its leverage. If instead it is lifted with straight legs and a bent back, it must be
supported almost entirely by the erector spinae muscles (fig. 2-5E). In this posture,
 Posture and the Back in Musical Performance 29

the downward force on the spine acts across the face of the intervertebral joints,
something they are not well designed to cope with. Lifting with the body twisted to
the side should also be avoided, as the stresses on the back under these circumstances
are about three times as great as if the body is held straight (Hall 2003). Many musical
instruments are either in themselves heavy or require robust protection for transport,
which adds significantly to their weight. Their awkward shapes may also tempt the
owners to adopt risky lifting postures when moving them in confined spaces. The dif-
ficulties cannot be eliminated completely if the player is alone, but simple strategies
such as adding wheels to cases, or carrying cases on the back using straps over both
shoulders so that the weight is supported symmetrically, will greatly reduce the risk
of injury. When moving pianos on their wheels, the main impetus should come from
pushing rather than pulling. When pushing against the piano, the force exerted on the
vertebral column acts to extend the spine. This is counteracted by the rectus abdomi-
nus muscle, which is both large and has a high mechanical advantage for flexing the
spine (fig. 2-5F). When pulling, the force exerted on the body flexes the spine and
this must be resisted by the erector spinae muscles, whose mechanical advantage, as
we have seen, is poor.

Playing
Many instruments can be played either standing up or sitting down. As we have seen,
the pressure on the disks is lower during standing, so where possible this posture should
be adopted for playing. While this is not always possible, for example, when playing
in an orchestra or other ensemble, it can at least be used during practice, which often
represents a period of several hours a day. Of course even when standing care must
be taken to adopt good posture. For example, there is often a tendency to lean forward
unnecessarily, which puts extra strain on the back and should be avoided. This may be
exacerbated by having the music stand set too low, or may reflect eyesight problems
that make reading music from a distance difficult. If the forward leaning is related to
the weight of the instrument, frequent short rests should be inserted into practice ses-
sions, and a conscious effort should be made to gradually build up the strength of the
erector spinae muscles in order to be able to maintain good posture for the duration
of the bouts of playing needed in performance. Tenseness in the muscles of the back,
shoulders, neck, and arms due to performance anxiety, worries about making mistakes,
or concern that the instrument may slip increase fatigue and have a negative influence
on posture. As we are often unaware of such chronic tension, it is necessary to make
a conscious effort to take a subjective inventory of tension at regular intervals during
playing. An optimum posture is one in which your grip on the instrument is sufficiently
firm but flexible, while the playing fingers are relaxed. The back should be supported
but not rigid, and the shoulders and neck relaxed. Great players make playing look easy
because they have found ways to make playing easy on themselves. Optimizing their
posture and eliminating unnecessary tension allows them to play close to the physical
limits of the body when necessary, and well within these limits at other times.
The postures adopted by performers on different instruments can be divided into
two broad categories: those that at least appear symmetrical and those that are clearly
Figure 2-5
 Posture and the Back in Musical Performance 31

of necessity asymmetrical. Relatively symmetrical postures are used by players of

most woodwind instruments, pianists, and percussionists. Those whose postures are
significantly asymmetrical include string players, guitarists, brass players, and flutists.
Instruments with playing postures that are intrinsically symmetrical pose fewer prob-
lems for the back; however, their players may quite unnecessarily adopt overtense or
asymmetrical poses that should be consciously avoided. If standing, try to start from
a stable posture with the weight taken equally on both feet. This does not mean that
you should be rooted to the spot; in fact, changing position frequently is important as
this ensures the load is not always carried by the same joints and muscles, but avoid
spending long periods with the weight mainly on one leg, as this requires the spine to
tilt to that side to bring the center of gravity over the foot. For wind players, extreme
forward flexion or side-to-side movements of the trunk will also interfere with breath
control by pushing the abdominal organs up into the thorax.
Among the woodwinds, the flute is particularly problematical for posture (fig. 2-
6A). There are numerous hand and wrist problems associated with this instrument,
which will be discussed in chapter 3, but playing posture may, in addition, cause
problems in the shoulder region (Dommerholt 2000). In order to allow the right arm
sufficient freedom to move, either the trunk or the neck (or both) must be rotated to
enable the mouth to lie parallel with the lip plate. Any tendency to allow the trunk
or neck to slump forward simply compounds the problem and should be avoided.
This not only puts unnecessary strain on the back but also interferes with breathing.
Students are often encouraged to hold the instrument horizontally at shoulder height,

Figure 2-5. A–C. Sitting posture. A. When we sit erect on a chair, the pelvis tends to rotate back-
ward (dashed arrow) because the downward force of the trunk falls behind the point of contact
of the pelvis with the seat (the ischial tuberosities). As a result, the degree of lumbar lordosis is
reduced compared to that in erect standing. B. On a kneeling chair where the angle between the
leg and the trunk is about 135°, the center of gravity of the trunk is brought into line with the
point of contact of the pelvis with the seat. This minimizes the force rotating the pelvis backward
and therefore the muscular effort needed to retain an erect posture. The degree of lumbar lor-
dosis lies between that of erect standing and sitting, close to the form it takes up in the neutral
state when no muscles are acting on it. C. In a reclining posture, the pelvis is strongly rotated
backward because the center of gravity lies well behind the ischial tuberosities. The normal
curvature of the lumbar spine is reversed (becomes kyphotic). D–F. Pushing, pulling, and lifting.
D. When lifting an object, the power should come from the large, strong postural muscles of the
leg, such as the quadratus femorus (which straightens the knee) and the gluteus maximus (which
extends the hip joint). The back is kept straight; the erector spinae muscles only serve to stabilize
it in this position so that the force acting on the body bears directly down onto the joints between
the vertebrae. E. Lifting with a bent back is more dangerous. The power from the lift must come
from the small, weak erector spinae muscles; the force on the spine acts across the plane of the
intervertebral joints. The disks are squeezed strongly at the front, making a prolapse at the back
that may cause compression of the spinal nerves or spinal cord. F. When moving a heavy object
across the floor, it is better to push it than to pull it. When pushing, the reactive force of the
object on the body extends the spine, causing it to bend backward. This is opposed by the strong
rectus abdominus, which forms the front wall of the abdomen and acts with a large mechanical
advantage on the vertebrae. When pulling, the reactive force flexes the spine forward. This is
resisted by the weaker erector spinae muscles, which have a poorer mechanical advantage.
Figure 2-6
 Posture and the Back in Musical Performance 33

Figure 2-6. A–C. Playing posture for the flute, violin, and guitar. (The postures are exaggerated
to make the details more obvious.) A. Of the wind instruments, the flute requires the most awk-
ward posture. The head and trunk are rotated to the left. The left arm is held across the body and
the right arm is raised out to the side, which fatigues the muscles of the shoulders. This is often
alleviated by tilting the trunk and/or the head to the right. B. A typical stance of violin players,
showing the rotation of the head and tilt of the neck and the raised left shoulder, which produces
a lateral deflection of the lumbar and thoracic spine. C. The guitar played in the seated position
with the left foot raised onto a rest. The pelvis is tilted so that the weight is shifted toward the
left ischial tuberosity. The shoulders also slope downward to the left, inducing a strong lateral
curvature of the spine. To keep the head vertical, the neck must be deflected to the right. D–F.
The effect of piano stool position on posture. The middle figure (E) shows the stool set at an opti-
mal distance, which allows the upper arm to hang close to vertical and the forearm to be nearly
horizontal when the fingers touch the keys. The sitting posture is fully erect with good lumbar
lordosis, minimizing the amount of muscular effort needed to maintain the posture. In D the
stool is set too close to the keyboard. To prevent the chest from projecting over the keyboard,
lumbar lordosis is increased. The elbows now lie farther back relative to the trunk. This restricts
arm movement, and the shoulders may have to be raised to keep the forearms horizontal, which
may also restrict breathing. F. Here the stool is too far from the piano. Because the upper arm is
extended, the entire weight of the arm must be supported by muscular effort, whereas when the
stool is set at an optimal distance this is only true for the forearm. The trunk is flexed toward the
piano, increasing the kyphosis of the thoracic spine, and the neck is extended. The normal lumbar
curvature is reduced and may even be reversed. More muscular effort is required to maintain this
posture as the center of gravity of the trunk is moved forward.

which requires that the upper arm be slightly elevated. This requires considerable
activity in the supraspinatus muscle (see chapter 3). Prolonged contraction of this
muscle can restrict its blood supply because the vessels within it are compressed
(Dommerholt 2000); as the muscle consequently fatigues, the end of the flute starts to
droop, the head is tilted to the right to maintain the correct angle with the head joint,
and tension can be caused in the sternocleidomastoid muscle. A compromise in which
the head is from the beginning allowed to tilt to the right by a small amount is used
by many professional players.
Though it may not be immediately obvious to the observer, playing the trumpet,
trombone, or French horn puts considerable asymmetrical strain on the shoulder and
trunk musculature. In the first two cases, the instrument is held up and supported for
long periods by the left arm. The problems this poses are exacerbated by the greater
weight of modern instruments designed with larger bells to produce a bigger sound.
The length of the trombone further serves to increase the leverage its weight can exert
and a high, tense left shoulder is a common result. While the horn is not held so high,
its weight (particularly if it is a double instrument) is considerable and can cause dif-
ficulties, especially for young players. One of the long-term physical consequences
of playing such instruments is that the muscles that pull the shoulder upward (e.g.,
the trapezius and to a lesser extent the scalenes and levator scapulae—see chapter 3)
are held under chronic tension during playing. The nerves running to the arm pass
between the scalene muscles and can become compressed at this point. Because the
shoulder and upper arms are extended forward in playing, the muscles that support
them in this position (pectoralis major and the anterior fibers of the deltoid) become
stronger. This tends to produce a shortening of the muscle fibers, pulling the humerus
34 Chapter Two

(the bone of the upper arm) forward in its socket. Muscles at the back of the shoulder
that oppose this action and which are, in any case, smaller remain weak as they are
used much less. This reduces the diameter of a small channel on top of the shoulder
through which passes the tendon of the supraspinatus muscle, which raises the arm
out to the side. Here it can become pinched, leading to shoulder pain in what is known
as rotator cuff syndrome (see chapter 3). One remedy is to strengthen these weak
muscles so that they draw the arm (and also the shoulder) backward to open up the
channel again. If the upper arm is allowed to hang vertically and the forearm is held
horizontally at a right angle to it, this can be achieved by rotating the arm outward and
the shoulders backward against the resistance provided by an elastic exercise band.
However, exercises for this type of shoulder problem must be chosen with care and
after a proper analysis of the problem, as strengthening the wrong set of muscles may
have the opposite effect to the one intended.
Asymmetric postures are unavoidable for players of stringed instruments. The con-
sequences of holding the violin or viola under the chin are several (Kapandji 2000;
Wynn Parry 1998). The left shoulder is often raised (hunched) and the chin pressed
down into the chin rest. This causes fatigue in muscles, such as the trapezius (see
chapter 3), that raise the shoulder. There is also often a tendency to rotate the left
shoulder forward. Together these postures cause the thoracolumbar spine to curve
to the right while the cervical spine is tilted to the left (fig. 2-6B). The cervical and
thoracic regions of the vertebral column will be also be rotated in opposite directions
in the horizontal plane. Over time the muscles on the left side of the trunk become
stronger at the expense of those on the right, influencing posture even when not play-
ing. The mild deformities of the spine that this produces in string players have been
recorded by Bejjani, Kaye, and Benham (1996), who found that the left shoulder was
chronically raised in many violinists and that harpists also had a particularly high
incidence of scoliosis (29 percent in this sample).
It is important to minimize the deviation of the vertebral column from the normal
erect position as much as possible. In the case of violin and viola players, the shoul-
ders should be kept in line with the hips so that there is no torsion of the trunk. The
shoulders should be kept down and the temptation to rotate the left shoulder forward
resisted. If the instrument feels unstable, the shoulder and chin rests need to be ad-
justed. For the most part, the head faces forward (so that the neck is not twisted) with
the instrument held out to the side. The music stand should therefore be straight in
front of the body. It is also important to realize that we are not all identical in size and
shape and that how the instrument is held and supported will vary between individu-
als. For example, the most ergonomically favorable angle at which the instrument is
held out to the left will depend on arm length (Fischer 1998). Shoulder and chin rests
are generally used to limit the degree to which the head must be tilted on the neck to
stabilize the instrument, and tailoring one to suit will need some experimentation, both
with different models and their fine adjustments.
The chin rest first appeared in the 1820s and its invention is usually attributed
to Louis Spohr (Rabuffetti et al. 2007). It was a response to the greater freedom of
movement that was required by the increased technical complexity of the nineteenth-
century violin repertoire; however, there is no evidence that its design, either then or
 Posture and the Back in Musical Performance 35

subsequently, was based on objectively determined ergonomic principles. With the

exception of performers who play baroque music on period instruments, the use of
chin rests today is almost universal. The origin of the shoulder rest is less clear, though
the use of a pillow under the instrument was mentioned in L’art du violon by Bail-
lot in 1834. Prior to this a chamois leather was often used to prevent the instrument
from slipping. The expected effect of using a shoulder rest on the muscles involved in
gripping the instrument has been confirmed using electromyography. It significantly
reduces the activity in the trapezius (which raises the shoulder) and the sternocleido-
mastoid (which rotates the head to the opposite side and tilts it upward) (Levy et al.
1992). Activity is increased in the anterior fibers of the deltoid, which helps raise the
arm out to the side, particularly when the arm is rotated outward as it may be when
holding the violin.
When chin rests and shoulder rests are used in combination, it appears that the na-
ture of the former is more significant than that of the latter in determining the degree
of force used to hold the instrument (Okner, Kernozek, and Wade 1997). Though the
design of the chin rest has a significant effect, it does not always follow the trend pre-
dicted. For example, Okner et al. found that a chin rest that was placed more centrally
on the instrument, which was predicted to show the lowest loadings, in fact had the
highest. There is currently insufficient information to know what aspects of chin rest
design have most effect on chin pressure. However, the study found that playing more
technically demanding music consistently led to the generation of higher pressures,
but whether this was due to a greater requirement for support or the result of tension
generated by psychological factors was not investigated.
Tall players with long necks often find that standard rests do not have a sufficient
range of adjustment for comfort (Okner, Kernozek, and Wade 1997), and this contrib-
utes to the fact that they have a high incidence of posture-related problems, including
spasms in the trapezius muscle, strain in other muscles of the neck, and pinching of
nerve roots where they emerge from the spine. The consequences of changing the
height of the shoulder rest on playing movements was examined in a group of fifteen
violinists by Rabuffetti et al. (2007), though the analysis did not take into account
the lengths of their necks. The height of the rest used could be adjusted over a range
of 40 millimeters. Raising the rest had the beneficial effects of reducing the leftward
rotation of the head and the elevation of the left shoulder, which was accompanied by
a reduction in activity in the trapezius muscle, and also reduced the internal rotation
of that shoulder. The violin was displaced forward and toward the right, but none of
the changes affected the players’ ability to maintain the correct angle between the
bow and the strings. Any postural changes required of the right arm took place at the
shoulder and so did not affect the wrist and hand posture used for holding the bow.
Increasing the height of the rest did have some negative consequences for shoulder
flexion and the degree of arm pronation; therefore, for any player the optimum rest
height remains a compromise.
As many commercial shoulder rests have only a limited range of adjustment, it may
be necessary to find specialist suppliers for these and for chin rests, or alternatively to
invest in a customized support. This can involve having the surface of the rests molded
to fit the player (Norris 2000) so that the force generated in holding the instrument is
36 Chapter Two

spread evenly over the entire contact area. The position of the chin rest on the instru-
ment may also have to be adjusted in relation to arm length to allow full and free use
of the bow. A discussion of some of these issues can be found in Lieberman (2000).
It is important to note that young players whose bodies are growing rapidly may need
regular adjustments in setup during their teenage years. Some of the postural conse-
quences that may arise from problems with gripping the violin or viola and the effect
of their resolution on postural symmetry are shown in animation 2-2. If you can access
the commercially available video recordings of David Oistrakh (e.g., in Beethoven’s
Fifth Violin Sonata), you will see an excellent example of an ergonomic playing pos-
ture. Holding the violin under his chin hardly disturbs the symmetry of his head and
shoulders, and the mobility of his head demonstrates that little tension is needed to grip
the instrument. His fingers are relaxed so that even in the most rapid passagework, he
appears to have plenty of time to make each movement.
Because of how the instrument is held, violin and viola players are susceptible to
pain at the jaw joint (the temporomandibular joint) and in those who start at a very
young age, the face may even become slightly asymmetrical as a result. A habit of
exerting greater pressure than necessary is sometimes associated with a clenching of
the teeth, which can lead to pain in the temporalis and masseter muscles that close the
jaw (see chapter 6). Remodeling and wear of the surfaces of the joint may result, as
well as excessive wear and even fracturing of the cusps of the molar teeth (Yeo et al.
2002). If it is consistent with the full use of the bow arm, moving the chin rest to the
center of the instrument, so that it lies over the tailpiece, is one way of reducing the
forces on the jaw (Blum and Ritter 1990). A common contributory factor to temporo-
mandibular joint problems is stress, and nonplaying consequences such as nocturnal
teeth grinding is often one result. As with many other problems encountered by musi-
cians, treatment should address the underlying cause as well as the symptoms. While
the use of plastic guards worn over the teeth to protect them from nighttime grinding
may be appropriate to treat the symptoms, the ultimate resolution of the problem may
also require modification of the shoulder and chin rest to reduce neck tension, as well
as postural retraining and stress management.
The postural problems posed by the double bass are quite considerable and merit
special mention. When standing, the player often has to lean to the right to allow the
bow arm low enough access around the side of the instrument to the strings. This can
be made a little easier by rotating the instrument to the right. Nevertheless, it is not
unusual to see players whose right shoulder slopes sharply downward, while the left is
almost horizontal. Furthermore, with such a large instrument, the height of the player is
a significant factor for posture. Short players must hold the upper arm much higher out
to the side to reach the top of the fingerboard, which leaves them prone to rotator cuff
problems in the shoulder (see chapter 3). They may also experience problems reaching
over the left side of the instrument to gain access to the lower end of the fingerboard.
Playing seated (which is common in orchestras) can introduce its own problems. A
player sitting symmetrically on the stool must lean far forward over the instrument in
a position which is almost calculated to cause vulnerability to disk prolapse. For this
reason it is more common to sit asymmetrically with one foot on the floor and one on
the spar of the stool or on a footrest in order to minimize forward leaning.
 Posture and the Back in Musical Performance 37

In the normal playing position for the classical guitar, the left foot is often placed
on a footrest so that the thigh can be raised to a suitable level to support the instru-
ment. The trunk may be tilted to the left so that the weight of the body is shifted onto
the left buttock, which in turn may lead to the right shoulder being raised and the left
lowered, and the neck inclined to the left to keep the head vertical. As a result, the
spine assumes an S-shaped curvature when seen from behind (fig. 2-6C). In addition,
the player may hunch forward over the instrument to see the frets more clearly or
because the back muscles are weak. This not only puts a strain on the back but also
puts additional strain on the neck, as the head must be supported by muscular effort.
It also results in the neck being bent backward to keep the head level, which tightens
the throat and restricts breathing. Much of this can be corrected by careful attention
to posture and by raising the music stand to a suitable level; however, the asymmetry
caused by raising the left foot remains. One option is to keep both feet on the ground
and place a firm, dome-shaped cushion on the left thigh to support and raise the in-
strument. This will help to reduce the lateral curvature of the spine. Other than this,
avoidance of postural problems is best promoted by taking frequent rests to stretch
and adjust posture during practice sessions (see comments by John Williams in Wynn
Parry 1998).

SEATING

As many musicians mainly perform while sitting down, the question of seating is an
important one, though it is only recently that it has begun to receive the attention it
deserves (Horvath 2002, 2004). Not only do the postures required for playing different
instruments vary considerably, but so do the dimensions of players. However, with the
exception of providing high stools for double-bass players, this is rarely reflected in the
seating made available to orchestras, which is usually in the form of a nonadjustable
stacking chair of a standard size. In a typical orchestra of about one hundred players, it
would not be surprising to find heights ranging from around five feet tall to well over
six feet. To start with a rather extreme example, if a seated player is very short and
cannot put the entire sole of the foot comfortably on the ground, significant pressure is
experienced in the soft tissues of the back or thighs, which soon become uncomfortable
(Keegan 1953). Tall players may have problems with the seat being too low, which
reduces the lordosis of the lumbar spine. For wind players, this can have the secondary
consequence of making it difficult to use full lung capacity. Occasionally one sees tall
orchestral players sitting on a pair of stacking chairs to counter this problem, but this
is a crude and unsatisfactory solution. For some instruments and players it is possible
to play while sitting with the lower part of the back resting against a lumbar support
(if the chair has one and the player has sufficient thigh length). Even if this is not the
preferred playing position, the possibility of using the lumbar support to rest between
bouts of playing provides considerable advantages. For the wind players in a symphony
orchestra, this is a significant proportion of the time spent sitting on stage. For string
players, who must spend most of the performance in the playing position, the question
of posture and seating is perhaps most acute. Cellists and bass players in particular may
38 Chapter Two

find themselves in forward-leaning postures for long periods, which puts considerable
strain on the back. Chairs that allow the flat surface of the seat to be tilted downward
toward the front can be an advantage for these (and some other) players for the reasons
outlined above. This posture can be adopted on normal chairs by sitting close to the
front edge so that the thighs can slope downward or by placing blocks under the chair’s
back legs, a form of customization often seen on the stage.
The ideal chair should have adequate padding, adjustable seat height and tilt, and
lumbar support. Is this an impossible dream? No—it is a description of the standard
seating used by office workers, the only difference being that the musician’s chair
should not swivel or run on casters unless chaos is to ensue on the platform. The rea-
son why office workers do not have to suffer the type of seating normally provided on
stage is that health and safety regulations prohibit it. Why should musicians have to
put up with less? There are indeed some chairs designed specifically with musicians in
mind that include (or claim to include) many such features. Unfortunately, the easiest
way to incorporate the degree of adjustability required is to use a design with a central
pillar. This means the chair can’t be stacked, which may not please the management.
Some stacking chairs designed for musicians have much less adjustability while others
do not deliver the levels of comfort they promise, so it is vital to try a variety of prod-
ucts with players of different sizes if you are considering these for your ensemble.
There is no single ideal sitting posture that can comfortably be maintained for long
periods of time regardless of whether it promotes what is regarded as optimal lumbar
lordosis. We frequently adjust our posture if the chair allows it, in order to redistribute
the weight between the various load-bearing surfaces of the body. This ensures adequate
blood flow to all of the tissues that are being compressed against the seat. This surface
should not therefore be shaped in a way that restricts movement. Some of the more
extreme “postural” designs such as kneeling stools may not allow much movement and
may transfer some of the weight onto the less fleshy parts of the leg and knee. They
may therefore not be comfortable for long periods. Sitting erect on a kneeling stool or
on one with a saddle-shaped seat to optimize lumbar lordosis requires both more bal-
ance and, initially at least, more muscle activity than a slumped position (Gandavadi
and Ramsay 2005), so being able to maintain an erect posture in a stool lacking lumbar
support requires practice and fitness of postural muscles. The same is true for sitting
erect at the front of the seat to achieve good posture, but the reward is a reduction in
the risk of back pain. The solution is not just in the design of the chair; greater postural
fitness will have other benefits for the quality of playing. In addition, a well-designed
chair should have a padded sitting surface made of a material to which the performer is
not liable to stick under warm conditions! While the chairs available during touring will
be beyond the control of the ensemble, an investment in good seating in the orchestra’s
home venue is likely to have considerable benefits for the long-term well-being of the
players, and will undoubtedly be reflected in performance quality. For touring, it may
be worthwhile to consider portable accessories, such as cushions (wedge-shaped or
otherwise), lumbar supports, and blocks for back chair legs, to cope with venues where
the seating is traditional or unpredictable (Horvath 2002).
In general, it is recommended that long periods of sitting be broken up by periodic
breaks of standing and walking, even if these last only a few minutes. This allows a
 Posture and the Back in Musical Performance 39

redistribution of the nucleus pulposus of the intervertebral disks (McGill 2002) and
relief for the compressed subcutaneous tissues of the buttocks. It can be followed by
mild stretching involving pushing the hands up toward the ceiling and inhaling deeply,
which gently and progressively extends the lumbar spine. These short breaks can eas-
ily be scheduled into rehearsals by understanding conductors who are likely to reap
their reward in improved responsiveness from the players.
With instruments such as the piano or harpsichord, the fixed position of the key-
board makes not only the height of the seat but also its distance from the instrument
critical. The design of the traditional piano stool at least makes it easy to regulate the
height, which should normally be set so that the forearms are close to horizontal or
slightly downward sloping during playing. If it is set too low or too high, the wrist
joint may be overflexed or overextended, which affects the freedom of movement of
the fingers (see chapter 3). Playing with hunched shoulders (whether developed as a
mannerism or as a result of incorrect seat height) should be avoided as not only does
this introduce tension in the shoulder and neck, but it also interferes with the recruit-
ment of the muscles in these regions when they are actually required for playing
(Brown 2000) (see animation 2-3). The stool should be placed in a position that allows
the keys to be reached comfortably with the body held erect (fig. 2-6D–F, animation
2-3). If the stool is too close to the keyboard, the pelvis is rotated forward and the chest
pulled back, leading to an exaggerated lumbar lordosis. This increases the muscular
effort required and the possibility of developing low back pain, and it may also restrict
breathing movements. It also means that the elbows are pushed backward so that the
hands do not extend too far over the keys, restricting their freedom of movement from
side to side. If the stool is too far from the keyboard, the pelvis tilts backward, straight-
ening out the lumbar curvature and increasing the thoracic kyphosis (Kapandji 2000).
The elbows straighten and greater muscular effort from the shoulders and upper arms
is needed to support the hands above the keys. This increases tension in the shoulders,
and because the center of gravity of the trunk is shifted forward, the erector spinae
muscles of the back must work harder to support the spine.
If you modify your playing posture to make it more favorable, it will almost cer-
tainly feel strange at first because you are so used to the original one, no matter how
awkward. Becoming acclimatized to the new posture will inevitably take time, but the
long-term benefits will be considerable.

EXERCISES AND POSTURAL TRAINING

Beyond the mechanical fixes afforded by good seating, it is important to be able to

make the most of the seat one is presented with by developing good posture. If one
watches the string sections of a full orchestra, the range of postures displayed is often
striking. Some look balanced and allow a natural movement of the body that appears
to be part of the expressive nature of the playing, while others look so rigid and un-
comfortable that it is hard to imagine that the sound being produced can be aestheti-
cally pleasing. This is particularly noticeable in student orchestras, where inexperi-
enced performers may be nervous or playing a second instrument with which they are
40 Chapter Two

not particularly experienced. However, it is often surprising to see that even among
good professional players, poor and risky postures are quite common. This may be ex-
acerbated by the setup of the instrument. For example, using an angled cello spike can
provide advantages to some players, but it should not be seen as an excuse to allow
the player to adopt a semirecumbent position! This will severely restrict movement of
the arm and trunk and lead to a high probability of back pain. Janet Horvath (Horvath
2002) and Barbara Paull (Paull and Harrison 1997) describe a variety of simple exer-
cises that can be used to relieve back tension and help build postural muscle strength
as well as some clever exercises that can be employed surreptitiously to stretch trunk
and other muscles even when on the platform. In rehearsal and practice, where the
constraints are much less, regular stretching and rest between bouts of playing should
be built into the sessions. These should not be seen as a luxury or fad but as a means
of optimizing performance and limiting the risk of injury. Prevention is the most sat-
isfactory and cost-effective way to treat injury.
Many players who are conscious of health issues have become interested in postural
development programs such as the Alexander technique, Pilates and Feldenkreis sys-
tems, tai chi, and yoga. While the benefits of these programs have for the most part not
yet been objectively assessed, the fact that they increase postural awareness, train im-
portant postural muscle groups, and increase flexibility is likely to be beneficial. There
is also some evidence from properly designed and evaluated trials that the Alexander
technique may have a positive effect on back pain (Ernst and Canter 2003). Many
postural therapies also help to reduce stress, which is usually associated with exces-
sive muscle tension, and contribute to reducing the impact of performance-related
anxiety (see chapter 10). Maintaining a fixed posture for prolonged periods can lead
to the development of some postural muscle groups at the expense of those that move
the joints in the opposite direction. This leads to an undesirable reinforcement of the
postural asymmetry, which can only be corrected by strengthening the weaker muscle
groups. General exercises that involve symmetrical activity in postural muscles (e.g.,
walking, swimming, or involvement in some other form of sport) and maintenance
of an acceptable body weight all contribute to good posture as well as provide many
other health benefits.

REFERENCES

Adams, M. A., N. Bogduk, K. Burton, and P. Dolan. 2002. The biomechanics of back pain.
Edinburgh: Churchill Livingstone.
Bejjani, F. J., G. M. Kaye, and M. Benham. 1996. Musculoskeletal and neuromuscular condi-
tions of instrumental musicians. Arch Phys Med Rehabil 77 (4):406–13.
Blum, J., and G. Ritter. 1990. Violinists and violists with masses under the left side of the angle
of the jaw known as “fiddler’s neck.” Med Prob Perform Artists 5 (4):155–60.
Brown, S. 2000. Promoting a healthy keyboard technique. In Medical problems of the instru-
mental musician, ed. R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Dommerholt, J. 2000. Posture. In Medical problems of the instrumental musician, ed. R. Tu-
biana and P. C. Amadio. London: Martin Dunitz.
 Posture and the Back in Musical Performance 41

Ernst, E., and P. H. Canter. 2003. The Alexander technique: A systematic review of controlled
clinical trials. Forsch Komplementarmed Klass Naturheilkd 10 (6):325–29.
Fischer, S. 1998. Technique and ease in violin playing. In The musician’s hand: A clinical
guide, ed. I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
Freemont, A. J., T. E. Peacock, P. Goupille, J. A. Hoyland, J. O’Brien, and M. I. Jayson.
1997. Nerve ingrowth into diseased intervertebral disc in chronic back pain. Lancet 350
(9072):178–81.
Gandavadi, A., and J. Ramsay. 2005. Effect of two seating positions on upper limb function in
normal subjects. Int J Ther Rehab 12 (11):485–90.
Hall, S. J. 2003. Basic biomechanics. 4th ed. Boston: McGraw-Hill.
Harrison, D. D., S. O. Harrison, A. C. Croft, D. E. Harrison, and S. J. Troyanovich. 1999.
Sitting biomechanics part I: Review of the literature. J Manipulative Physiol Ther 22
(9):594–609.
Horvath, J. 2002. Playing (less) hurt: An injury prevention guide for musicians. Minneapolis,
Minn.: J. Horvath.
———. 2004. Stressing prevention. Symphony (March–April).
Kapandji, A. I. 2000. Anatomy of the spine. In Medical problems of the instrumental musician,
ed. R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Keegan, J. J. 1953. Alterations of the lumbar curve related to posture and seating. J Bone Joint
Surg Am 35-A (3):589–603.
Levy, C. E., W. A. Lee, A. G. Brandfonbrener, J. Press, and A. E. Levy. 1992. Electromyo-
graphic analysis of muscular activity in the upper extremity generated by supporting a violin
with and without a shoulder rest. Med Prob Perform Artists 7:103–9.
Lieberman, J. L. 2000. The importance of setup. The Strad (May/June).
McGill, S. 2002. Low back disorders: Evidence-based prevention and rehabilitation. Cham-
paign, Ill.: Human Kinetics.
Norris, R. 2000. Applied ergonomics. In Medical problems of the instrumental musician, ed.
R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Okner, M. A. O., T. Kernozek, and M. G. Wade. 1997. Chin rest pressure in violin players:
Musical repertoire, chin rests and shoulder pads as possible mediators. Med Prob Perform
Artists 12:112–21.
Paull, B., and C. Harrison. 1997. The athletic musician: A guide to playing without pain. Lan-
ham, Md.: Scarecrow Press.
Rabuffetti, M., R. M. Converti, S. Boccardi, and M. Ferrarin. 2007. Tuning of the violin-
performer interface: An experimental study about the effects of shoulder rest variations on
playing kinematics. Med Prob Perform Artists 22 (2):58–66.
Richardson, C. A., and G. A. Jull. 1995. Muscle control—pain control. What exercises would
you prescribe? Manual Therapy 1:2–10.
White, A. A., and M. M. Panjabi. 1990. Clinical biomechanics of the spine. 2nd ed. Philadel-
phia: Lippincott.
Wynn Parry, C. B. 1998. The interface. In The musician’s hand: A clinical guide, ed. I. Win-
spur and C. B. Wynn Parry. London: Martin Dunitz.
Yeo, D. K., T. P. Pham, J. Baker, and S. A. Porters. 2002. Specific orofacial problems experi-
enced by musicians. Aust Dent J 47 (1):2–11.
 Chapter Three

The Shoulder, Arm, and Hand

Structure and Problems

The upper limb is capable of a great range of movements, which is a reflection of its
structure, the nature of its joints, and the action of its muscles. A knowledge of the
anatomy of the upper limb and how it works is important to instrumental musicians
for a variety of reasons. More than any other part of the body, it is the movements of
the arm and the hand that are used to create music. Just as much as your instrument,
these are the tools of your trade. By understanding how the movements are achieved,
as well as their intrinsic limitations, the player will be better able to optimize per-
formance without being tempted to demand what is not possible. Because of the
range and repetitive nature of the movements required, the arm, hand, and shoulder
are the most frequent sources of the pain and discomfort that can limit or even end
the ability to play effectively. An appreciation of what may go wrong and how to
reduce the risk of breakdown is the greatest assurance a player can have of a long
and trouble-free career. The risk of injury cannot be eliminated entirely, however,
so this chapter also explains the origin of some of the conditions that are most likely
to limit performance. Their symptoms are described, together with the typical forms
of treatment they may require. Among these symptoms are several that the player
should recognize as warning signs that must not be ignored. This knowledge should
provide the player with the confidence to seek help promptly and enable him or her
to communicate effectively with the doctor or therapist. This chapter will also discuss
some of the difficulties that may be experienced in diagnosing the primary source of
an injury and promote an appreciation of the importance of following any prescribed
treatment conscientiously.

ANATOMY OF THE UPPER LIMB

Most of what is covered in this chapter should be understandable without any previ-
ous knowledge of anatomical terminology (those of you who have this knowledge
already may want to grit your teeth at this point!). In many places I have added the
technical terms in parentheses as these will be useful if the reader wishes to explore

42
Figure 3-1. A diagram showing the anatomical names for the movements of the arm and hand.
44 Chapter Three

any of the topics further. (For example, such information will be necessary to cope
with many of the references at the end of the chapter.) Many of these terms are used
because everyday language either does not provide an equivalent or cannot be used to
describe important concepts with sufficient precision. This is nowhere more true than
in descriptions of the movements of joints. For this reason the appropriate terms are
explained in figure 3-1.

The Muscles and Joints of the Shoulder

The arm is relatively loosely attached to the trunk. Where two bones move against
each other at a joint, they are said to articulate. The arm articulates with the shoulder
blade (scapula), which in turn is braced against the breastbone (sternum) by a long
lever formed by the collarbone (clavicle) (fig. 3-2). The collarbone is relatively nar-
row and provides the only bony connection between the arm and the trunk. If we fall
on an outstretched arm it takes the full weight of the body, which is why it is prone
to fracture under these circumstances. The shoulder blade is otherwise attached to the
trunk only by muscles, which run both to the vertebral column (the spine) of the back
and the neck, and to the chest wall (figs. 3-3, 3-4). The advantage of this arrangement
is that it allows the shoulder blade to slide and rotate over the back of the ribcage,
contributing to the great freedom of movement of the arm.
The shoulder blade has a complex shape (fig. 3-2A, B) but is basically a triangular
plate with the apex pointing downward. The corner that points toward the arm ends
in a flat oval plate that forms the socket (the glenoid) against which the upper bone of
the arm (the humerus) articulates. A small bony process (the coracoid), shaped rather
like a thumb, pushes forward from the glenoid and several muscles in the arm attach to
it, the most familiar of which is the biceps. At the back of the shoulder blade, a long,
ridgelike spine runs horizontally near the top edge. The spine ends in a flattened plate
(the acromium) that lies horizontally above the shoulder joint, where the collarbone
attaches to it. You can identify the acromium by running your finger along the col-
larbone to where it ends at the point of the shoulder. The largest muscle holding the
shoulder blade in place is called the trapezius (fig. 3-3D). It has a triangular shape with
its longest edge attached to the vertebral column and the back of the skull, and also to
a ligament (the nuchal ligament) that runs between them. The muscle fibers run from
here to the spine of the shoulder blade. The upper fibers run obliquely downward,
forming the slope of the shoulders, and the middle fibers run horizontally, while the
lower fibers run obliquely upward across the back. Consequently, depending on which
fibers contract, the muscle can either raise the shoulder blade (as in shrugging), pull
it downward, or pull it inward toward the spine. Underneath the trapezius are shorter
muscles (the rhomboids) that pass from the vertical edge of the shoulder blade to the
vertebral column, pulling the blade inward and upward. In this they are aided by the
levator scapulae muscles, which pass more steeply upward to the vertebrae of the neck
(fig. 3-3B). The muscles that support and move the shoulder blade are often a site of
tension when we are under stress and are therefore a common target for massage. When
the arm is raised, the point of the shoulder is rotated upward by the shoulder muscles,
and so these are heavily used by string players. The shoulder blade is also attached to
Figure 3-2. The skeleton of the shoulder seen from the front (A) and the back (B). This shows
how the humerus of the arm articulates with the shoulder blade. The structures named in pa-
rentheses are parts of the shoulder blade. The arm and shoulder are only attached to the body
directly through the joint between the collarbone and the breastbone. The shoulder blade is
otherwise supported only by muscles suspending it from the ribcage, spine, and skull. C, D: The
bursa of the shoulder joint, revealed through the transparent deltoid muscle, as seen from the
front (C) and back (D). The bursa (which is like an uninflated balloon containing a film of synovial
fluid) allows the deltoid to slide smoothly over the underlying tissues, and as it extends under the
acromium contributes to the smooth movement of the supraspinatus and infraspinatus tendons.
46 Chapter Three

the rib cage by two other muscles. One is called the serratus anterior, and this runs from
the side of the upper ribs, wrapping around the chest wall under the shoulder blade, to
its vertical edge, which lies almost parallel with the vertebral column (figs. 3-3B and
3-4B). This muscle holds the shoulder blade against the back of the rib cage. The other
muscle, the pectoralis minor, attaches the coracoid process to the front of the rib cage
(fig. 3-4D) and assists the action of the serratus anterior.
The shallowness of the shoulder socket (glenoid) allows the arm a wide range of
movement, but at the price of instability. For this reason, the top of the humerus must
be held in position by muscles. Four muscles are involved (see animation 3-1). Two
of these are found on the back of the shoulder blade on either side of its horizontal
spine. One sits in the groove above the spine and is therefore called the supraspinatus.
This raises the arm out to the side, a movement called abduction. The muscle below
the spine (the infraspinatus) rotates the arm outward around its vertical axis (fig. 3-
3C). The third muscle (the teres minor) is small and runs from the lateral edge of the
shoulder blade to the back of the humerus, while the fourth (subscapularis) is a large
triangular muscle that has its origin on the flat front surface of the shoulder blade. This
also attaches the humerus and rotates the arm inward. The tendon of the supraspinatus
must pass through a narrow gap that lies beneath the acromium to reach the humerus,
and here it is vulnerable to being pinched if any of the tissues in this region become
swollen (figs. 3-2, 3-4C, animation 3-2). This is a potential source of problems for
string players in particular.
The shoulder joint is of the ball-and-socket type. The upper end of the humerus
ends in a smooth hemispherical surface that allows the arm to rotate in many direc-
tions against the other face of the joint, which belongs to the shoulder blade. As
with most moving joints, the surfaces that slide over one another are covered with a
smooth, glassy, hard-wearing type of cartilage (hyaline cartilage—animation 3-1). In
osteoarthritis, the surface of the cartilage becomes broken up and rough, restricting
movement and causing pain. The joint is what is called a synovial joint, meaning that
it is enclosed in a fibrous capsule or sac containing a lubricating fluid (synovial fluid)
(animation 3-1). The socket of the shoulder joint is very shallow and the capsule quite
loose. This allows the arm great freedom of movement, but this comes at a cost, as
the joint is prone to dislocation. The stability of the joint is maintained largely by the
broad, flat tendons belonging to the four muscles just described—supraspinatus, in-
fraspinatus, teres minor, and subscapularis. These fuse with the joint capsule as they
attach to the humerus just behind the ball-like articular surface that lies within the
joint. The tendons therefore run behind, above, and in front of the joint, but leave the
region below it (the armpit or axilla) unsupported (figs. 3-3C, 3-4C), which is why
dislocation of the shoulder almost always occurs in the downward direction. Together
the tendons form what is called the rotator cuff, and by acting in a concerted fashion,
their muscles hold the head of the humerus in place (animation 3-1). Acting individu-
ally, they can rotate the arm around its long axis or contribute to holding it away from,
or pulling it toward, the body.
Passing over the rotator cuff is the deltoid muscle, which arises from the shoulder
blade (both from its spine and the acromium) and from the collarbone. The other end
attaches to the lateral side of the upper part of the humerus. It therefore sits in a similar
Figure 3-3. Muscles attaching the shoulder blade to the back and to the arm. The arrows indi-
cate the directions in which the muscles can move the shoulder blade. A. The skeleton. B. The
rhomboids and levator scapulae pull the shoulder blade upward, while the serratus anterior holds
it against the back of the chest. C. The rotator cuff muscles attach the shoulder blade to the arm.
Their tendons deepen the socket of the joint to make it more stable. D. The muscle fibers of the
trapezius fan out from the spine of the shoulder blade to the spine and the base of the skull. The
upper fibers pull it upward, the middle fibers pull it inward, and the lower fibers pull it down-
ward. The deltoid muscle links the spine of the shoulder blade to the side of the upper arm, which
it helps to raise, while the latissimus dorsi pulls the raised arm downward.
48 Chapter Three

Figure 3-4. Muscles of the shoulder seen from the front. The arrows show the action of the
muscles on the shoulder blade. A. Skeleton of the shoulder. B. The attachment of serratus ante-
rior to the ribs. C. Muscles attaching the humerus to the shoulder blade. D. The pectoralis minor
muscle, which pulls the coracoid process of the scapula toward the ribs.

position to an epaulet. This muscle raises the arm out to the side (abduction) and there-
fore aids the supraspinatus muscle. To enable the deltoid to slide smoothly over the
rotator cuff tendons and the upper end of the humerus, it is separated from them by a
thin-walled sac called a bursa (this is a Greek word meaning bag or purse) (fig. 3-2C,
animation 3-2). The bursa extends under the acromium and also facilitates the smooth
movement of the supraspinatus tendon within its channel. Normally the bursa contains
a small amount of fluid, which allows the upper and lower surfaces of the sac to slide
 The Shoulder, Arm, and Hand 49

smoothly over one another as the deltoid and supraspinatus muscles contract, which
they often do together as they have similar actions. There are several other bursae
around the joint that contribute to the smooth movement of shoulder muscle tendons.
If the surfaces do not slide smoothly and become stuck to one another (bursitis), pain
and restriction of movement occurs (see below).
In most synovial joints there are no tendons running within the capsule, but the
shoulder joint is unusual in this respect as a tendon belonging to the biceps muscle does
enter the cavity. The biceps has two parts or heads (hence its name). As mentioned ear-
lier, one is attached to the bony coracoid process of the shoulder blade that lies outside
the joint; the other sends a narrow tendon into the capsule to reach its attachment site
on the upper edge of the articular surface of the socket (the supraglenoid tubercle). The
lower end of the biceps ends in the forearm, where it is attached by another tendon to
a bone called the radius. The tendon is also attached to a sheet of fibrous tissue (deep
fascia) that wraps around the muscles of the forearm. The biceps muscle has a number
of functions; it flexes the shoulder and the elbow joints, but because of its attachment to
the deep fascia it also helps to supinate the forearm (rotating the forearm so that when
the elbow is straight, the palm faces forward). Supination is required for the playing of
many instruments, for example, the left arm of the violin, viola, and flute.

The Muscles and Joints of the Elbow

The upper arm has only a single bone (the humerus—fig. 3-2), but at the elbow this
articulates with the two bones of the forearm, the radius (which runs down the thumb
side of the forearm) and the thinner ulna on the little-finger side (figs. 3-5, 3-6). The
elbow joint is quite a complex affair. The main part is a hinge linking the humerus
with the ulna; this allows movement in only one plane, but the joint also includes an
arrangement that allows the radius to rotate. The ulna projects behind the end of the
humerus (the point of the elbow). This provides an attachment site for the large muscle
at the back of the arm, the triceps, which straightens (extends) the elbow (fig. 3-5A).
The triceps has three heads, two of which arise from the humerus and one from the lat-
eral edge of the shoulder blade, so like the biceps it also acts on the shoulder joint. The
triceps is the only major muscle that extends the elbow; however, the biceps is aided
in flexing the elbow by a number of other muscles including the brachialis, coracobra-
chialis, and brachioradialis, which run underneath and alongside it (fig. 3-5B, C).
The radius ends at the elbow in a round, disk-shaped head that articulates with the
lower end of the humerus. Beneath the head is a narrow neck that is held in place by a
tendinous ring (the annular ligament). This allows the radius to rotate around the ulna.
If we hold the arm straight with the palm of the hand facing forward (supination), the
radius and ulna lie parallel (fig. 3-6). When the palm of the hand is rotated to face back-
ward (pronation), the ulna remains fixed and the radius rotates across it. This is brought
about by pronator muscles that run between the radius and ulna. Keyboard instruments
are played with both forearms pronated (i.e., with the thumb pointing toward the body),
while the violin, viola, and flute are played with the right forearm pronated. Other
woodwind and brass instruments such as the trumpet are played with the forearms
in an intermediate position. Like the shoulder, the elbow joint is a synovial one and is
50 Chapter Three

Figure 3-5. Muscles of the upper arm. A. View from the back of the triceps, a muscle that acts
on the shoulder and straightens the elbow. B. View from the front showing the coracobrachialis,
which acts on the shoulder joint to flex the arm. The brachialis and brachioradialis flex the elbow.
C. The biceps acts on both the shoulder and the elbow.

associated with several bursae. One that lies beneath the skin at the point of the elbow
is the most likely to become inflamed, for example, as a result of prolonged leaning on
a desk (draftsman’s elbow).

The Muscles and Joints of the Wrist

You may think of the wrist mainly as a hinge joint allowing up and down movements
of the hand at right angles to its flat surfaces (as in waving); however, the curvature of
the articular surfaces of the joint also allows movements from side to side in the plane
of the hand. These movements are called radial or ulnar deviation (animation 3-5).
Ulnar deviation is particularly important for violin and viola players as the left hand
moves up the fingerboard toward the bridge. Pianists use both radial and ulnar devia-
tion as their fingers run sequentially up or down the keyboard. The wrist is moved by
two flexor and three extensor muscles running down each side of the forearm (figs. 3-
8, 3-10). When both flexors or all of the extensors are active, the hand is simply drawn
back (extended) or pulled forward (flexed), as in waving. Radial or ulnar deviation is
caused when the flexor and extensor muscles on the same side contract together.

The Muscles and Joints of the Hand

The skeleton of the palm of the hand is made up of eight small bones (the carpals) to-
gether with five longer ones (metacarpals), each supporting one of the digits (fig. 3-7).
 The Shoulder, Arm, and Hand 51

Figure 3-6. Supination and pronation of the forearm. If we stand with the arms straight and the
palms facing forward, the two bones of the forearm (radius and ulna) run parallel. In this position,
the thumbs point away from the body and the forearm is said to be supinated. Panel A shows
this from the back of the arm and B, from the front. If the hand is then turned so that the palm
faces the back and the thumbs point toward the body, the radius rotates across the ulna and the
forearms are said to be pronated. The muscles responsible for these movements are indicated in
larger typeface.

The small carpal bones are difficult to feel as separate entities, but the metacarpals can
be easily appreciated by running a finger down the back of the hand from each finger
or from the thumb. The radial and ulnar bones of the forearm articulate with three
of the carpal bones. These are firmly bound by ligaments to the other carpal bones
and to the metacarpals to form a fairly rigid framework for the palm. The heads of
the metacarpals are also bound tightly together at the base of the fingers. The fingers
themselves each contain three bones (phalanges). The metacarpal bone of the thumb
has much more freedom of movement than those of the fingers. This is necessary to
allow the thumb (which has only two phalangeal bones) to be brought into opposition
with the other fingers to enable efficient gripping. We need not concern ourselves with
the names of the carpal bones and their joints, but the names of the individual finger
joints are important for delving deeper into the mechanism of finger action and its
pathology. The knuckle is the metacarpophalangeal joint (sometimes abbreviated to
MP), which is capable of flexion/extension, abduction/adduction, and circumduction
(movement in an arc). The other two joints of the finger are hinge joints at which only
flexion/extension is possible. The one nearest the knuckle is the proximal interphalan-
geal joint (PIP) and the last one, the distal interphalangeal joint (DIP).
52 Chapter Three

Figure 3-7. Bones and joints of the hand. The skeleton of the base of the palm is formed from
eight small carpal bones. The rest of the palm is made up of the five long metacarpals, while the
fingers and thumb are supported by the phalangeal bones. The joint at the knuckle is called the
metacarpophalangeal joint. The other two joints of the fingers are interphalangeal joints: the
middle joint is the proximal interphalangeal joint (PIP) and the last one, the distal interphalangeal
joint (DIP). Particular care is needed in treatment of injuries to the PIP to ensure that the full
range of movement is retained.

Many players of stringed instruments believe that playing increases finger span and
that this can be encouraged by stretching exercises; however, only a single study has
examined this question rigorously. It compared the difference in span in the right and
left hands, both between adjacent fingers and the total span between the index and
little finger. The subjects were sixty-nine cellists, fifty-two guitarists, and eighty-three
nonplaying control subjects, all right handed (Kloeppel 2000). The spreadability of the
gap between several of the fingers was greater in the left hand in all three groups. The
reason for this result in the control group is not known, but it has been speculated that
specialization of the right hand for gripping may be a factor. The difference between
the total finger span of the two hands was statistically significant only for cellists and
appears to be the result of an increased flexibility at the metacarpophalangeal joints.
However as the difference was only a few millimeters, its significance for playing
would be quite minor. Differences in span were not associated with the age at which
the instrument was taken up, though there was a significant correlation with current
practice intensity.
The organization of the muscles used to move the fingers (or digits) is quite com-
plicated and is a reflection of the degree of control we have over the individual joints.
The muscles can be divided into two major groups, those that lie in the forearm (figs.
3-8, 3-10) and those that are found in the hand itself (figs. 3-9, 3-11, 3-12). The
 The Shoulder, Arm, and Hand 53

Figure 3-8. Muscles on the inner side of the forearm that flex the fingers and wrist. These
muscles are shown in layers to make them easier to understand. A. The deep flexor muscles of
the fingers and the long flexor of the thumb. The tendons of the deep flexors run all the way to
the last joint of each finger. B. The superficial flexor muscles of the fingers. The tendons of the
superficial flexor muscles do not act on the last finger joint. C. Flexors of the wrist.

forearm muscles are of course much larger and provide the power for finger move-
ment and grip, whereas the muscles within the hand provide a more sophisticated and
independent control over finger movement. We will start by considering the forearm
muscles that flex (curl) the fingers. These of course lie on the side of the forearm that
is continuous with the palm.
There are two layers of flexor muscles for the fingers, each with different actions.
The more superficial layer (flexor digitorum superficialis) arises mainly from the ul-
nar (little finger) side of the humerus just above the elbow but is also attached to the
ulna and radius (fig. 3-8B). Though named as if a single muscle, it is actually a fusion
of several muscles that can be controlled separately, giving the individual fingers a
considerable (though not complete) independence of movement. Superficial flexor
tendons run down each of the four fingers, where they split into two and insert into
the sides of the middle phalangeal bone (fig. 3-9C). This means that they flex only the
knuckle (metacarpophalangeal joint) and the first (proximal interphalangeal) joint in
the finger. To flex the last joint of the finger (the distal interphalangeal joint) requires
the action of the deep flexor muscles (flexor digitorum profundus), as demonstrated
in animation 3-3. This muscle arises mainly from the ulnar bone and again, one ten-
don runs into each of the four fingers, underneath the superficial flexor tendon (fig.
3-8A). Where the superficial tendon splits, the deep tendon emerges to attach to the
last finger bone. The part of the muscle supplying the index finger is better separated
Figure 3-9. Muscles and tendons on the palmar side of the hand shown in layers. A. The palmar
interosseus muscles. The direction in which they move the fingers is shown by the arrows. Also
visible is the tendon of the long flexor of the thumb. This is sometimes linked to the deep flexor
tendon of the index finger (Linburg-Comstock syndrome). B. The tendons of the deep flexor
muscles of the fingers and their attached lumbrical muscles. C. At each edge of the hand are
short muscles that act on the thumb and little finger to cause flexion, abduction, and opposition.
The flexor retinaculum, which holds the flexor tendons down at the wrist, is also visible. D. The
tendon sheaths (tenosynovia), which help to ensure free movement of the tendons under the
retinaculum and the fibrous bands that hold the tendons down onto the fingers.
 The Shoulder, Arm, and Hand 55

Figure 3-10. Muscles on the back of the forearm that extend the fingers and wrist. The muscles
are shown in layers. A. The deepest muscles are those running to the thumb (two extensors and,
farthest toward the radial edge of the arm, an abductor). Also in this layer is the extensor of the
index finger. B. The next layer contains the common extensor for the four fingers plus the ad-
ditional and separate extensor of the little finger, which contributes to its independence of move-
ment. C. Extensors of the wrist.

than that to other fingers, giving the index finger a greater degree of independent
movement. When a muscle contracts, it moves all of the joints between its origin
and the place where its tendon is attached, so the deep flexor, in addition to moving
the distal phalangeal joint, will flex the other joints of the finger and also contribute
weakly to flexion of the wrist. Consequently, if the intention is to move only some
of the joints, the movement of the others will have to be resisted by the action of an-
tagonistic muscles. In this case, for example, wrist extensor muscles could be used to
keep that joint straight.
The arrangement of the finger extensor muscles that lie in the forearm is rather
different (fig. 3-10). There is only a single layer of common extensor muscle whose
tendons run to the four fingers; however, in addition, there are entirely independent
extensor muscles for the index and little fingers. All three muscles arise from what
is called the common extensor origin on the radial (thumb) side of the humerus just
above the elbow joint. The common extensor tendons become flattened where they
attach to the three bones of the fingers. At the base of the first phalangeal bone of the
finger, just beyond the knuckle joint, the tendon expands into a hood to which some
small muscles of the hand also attach (figs. 3-11C, 3-12B, animation 3-3). On the back
of the hand, the separate tendons of the common extensor muscle are linked to each
other by a variable pattern of fibrous bands. In addition, some tendons run to a point
Figure 3-11. Muscles and tendons on the back of the hand shown in layers. A. Dorsal interosseus
muscles. The direction in which they move the fingers is shown by the arrows. B. Tendons of the
extensor of the index and little fingers, and of two extensors and a long abductor of the thumb.
C. Tendons of the common extensor of the fingers. These are linked together by fibrous bands on
the back of the hand that restrict their independence of movement. D. The tendon sheaths and
the extensor retinaculum.
 The Shoulder, Arm, and Hand 57

midway between the base of two fingers, where they divide to send one onto each
finger. The tendon to the third finger is usually particularly tightly attached to that
of the little finger. The consequence of this enslavement between fingers can clearly
be observed in studies of the finger movements made by violinists (Wiesendanger,
Baader, and Kazennikov 2006; Baader, Kazennikov, and Wiesendanger 2005). Often
the enslaved finger does not make contact with its string; however, depending on the
passage, this may not be critical and indeed, it may be possible to place this finger on
a nonsounding string at a point that corresponds to the next note to be played upon it.
A greater independence of the little and index fingers is, however, ensured by their
separate extensor muscles. It has been proposed that cutting some of the fibrous bands
linking the extensor tendons may benefit pianists by reducing the tension and thus
limitation of these tendons, but there is absolutely no evidence to support this. One of
the main factors in this is the limited degree to which the individual parts of the com-
mon extensor muscle controlling each tendon can be activated independently.
The ability to control the movement of the fingers individually is aided by a number
of small muscles lying deep within the hand. Attached to the surfaces of the metacar-
pal bones are two sets of interosseus muscles, one in the palm and one on the back
(dorsal surface) of the hand (figs. 3-9, 3-11, 3-12). Their tendons insert into the side of
the first of the finger bones (the proximal phalanges) at the extensor tendon expansion.
The dorsal interosseus muscles spread the fingers, and the palmar interosseus muscles
draw them together. Another set of muscles deep within the palm are called the lum-
bricals because of their long, round, wormlike appearance. These arise from the deep
flexor tendons and run to the extensor tendon expansion. Their action is complex, as
the activity of the deep flexor muscles moves the tendons to which they are attached.
Their main role is simultaneously to flex the knuckle and straighten the fingers by
pulling on the extensor tendon expansions (fig. 3-12, animation 3-3). In this they are
aided by the interosseus muscles. When the lumbrical and interosseus muscles are
paralyzed, the hand assumes a clawlike form with the fingers curled in (i.e., the joints
between the finger bones remain permanently flexed).

The Muscles in the Thumb and Little Finger

The little finger is controlled by some additional muscles within the hand that arise
from the carpal bones and form a fleshy mass at the edge of the palm called the hy-
pothenar eminence (fig. 3-9C). One of these is an abductor, whose role is similar to
a dorsal interosseus muscle, drawing the little finger outward when the fingers are
spread. In addition, there is a short flexor muscle that helps to curl the little finger into
the palm. Finally, there is an opponens muscle that pulls the little finger over to meet
the thumb to allow a prehensile grip.
The independent movement of the thumb is also important for gripping, and the
long muscles that control it are quite distinct from those that move the fingers. On the
front of the forearm there is a single long flexor of the thumb (flexor pollicis longus)
(fig. 3-8A). On the back there are two extensor muscles (extensor pollicis longus and
brevis) and an abductor muscle, which draws the thumb away from the hand (fig.
3-10A). The tendons of these muscles pass down the side of the thumb. Two run
Figure 3-12. The organization and actions of the lumbrical muscles. A. The lumbricals arise from
the deep flexor tendons but end in the extensor tendon expansion, forming a broad, hoodlike flap
that attaches to the side of the finger just after the knuckle (B). C. The flexor muscles are active,
curling the fingers, but the lumbrical muscle is relaxed. D. The lumbrical has contracted, pulling
on the extensor tendon expansion. This straightens (extends) the finger, while the knuckle joint
is flexed.
 The Shoulder, Arm, and Hand 59

together, so when the thumb is stretched outward and backward, only two cords are
visible, one representing the tendon of the long extensor and the other, the tendons of
the abductor and the short extensor. The space between them is rather quaintly known
as the “anatomical snuff box” because when snuff was a popular way to take tobacco,
a pinch would be placed in this groove prior for sniffing. The thumb is also moved by
muscles that form a mass that runs along the edge of the palm at its base (the thenar
eminence). As with the little finger, this contains a short flexor, a short abductor, and
an opponens muscle. In addition, there is a muscle (an adductor) in the web of skin
between the thumb and index finger that pulls the thumb inward toward the palm (fig.
3-9C).

Tendon Movements in the Wrist and Hand

Many of the arm and hand problems of musicians involve places where tendons and
nerves pass through restricted channels. This is most acute at the wrist. If the wrist is
strongly flexed or extended, the long tendons that run to the fingers from the forearm
will try to take the shortest route. If this was not restricted, they would rise above
the surface of the wrist joint, raising the skin like a tent. In order to prevent this, the
tendons are held down by strong bands of tissue called retinacula, which wrap around
the wrist like a sweatband (figs. 3-9C and 3-11C, animation 3-4). The channel under
the flexor retinaculum at the base of the palm is known as the carpal tunnel. Simi-
lar, smaller bands hold the tendons down onto the surface of each finger (fig. 3-9D,
animation 3-4). The tendons are therefore confined to narrow channels beneath the
retinacula. In order for them to be able to move freely, they are wrapped in tendon
sheaths (figs. 3-9D, 3-11D) composed of two layers of synovial membrane forming a
double cylinder (fig. 3-13A). The inner layer is wrapped around the tendon and, lubri-
cated by the synovial fluid, this slides smoothly under the outer layer. The technical
term for these tendon sheaths is tenosynovia. If the synovial membranes or the tendon
become inflamed (tenosinovitis), movement becomes restricted and painful. This can
create a vicious circle as the swelling caused by the inflammation results in more fric-
tion, leading to further damage and inflammation. The tightness of the spaces under
the retinacula depends on the joint angle. For musicians, this is again most critical at
the wrist. If the wrist is held straight, the cross-sectional area of the spaces beneath
the retinacula are at their largest. If during playing the wrist is held strongly flexed or
extended, or deviated in the ulnar or radial direction, the space is more restricted and
the tendons are pushed together (see animation 3-5). In addition, when the wrist is
deviated from the neutral position, the tendons of one set of muscles will be stretched,
and this may limit the angle through which the finger joints can be moved by their
antagonists. To demonstrate this, hold the wrist straight and make a tight fist. Now
slowly flex the wrist, gradually forcing the hand farther downward; the fingers start to
uncurl as the extensor tendons become taut. In other words, with increasing wrist flex-
ion, the ability to flex the fingers is reduced. Clearly, then, if the wrist is in the neutral
position, not only is the ability to move the fingers maximized but the movement of
the tendons under the retinaculum is less likely to cause irritation or damage to it or
its sheath due to friction. The player should therefore aim to develop a wrist posture as
Figure 3-13. The tendon sheath and trigger finger.
 The Shoulder, Arm, and Hand 61

close to neutral as is compatible with the playing of the instrument. In some cases this
will be a simple question of playing posture, but if the difficulties cannot be resolved
in this way, modifications to the instrument may make the difference between being
able to continue playing and being forced to stop.

Limitations to Independent Finger Movement

The hand has evolved primarily for grasping, and though in man and other primates
it has the added sophistication that the thumb can be opposed to the other fingers, the
demands made of it in instrumental performance can push its design specifications to
their limits. This is particularly marked in keyboard playing, though it also applies
to some extent to string and woodwind players. The classical textbook description of
the actions of the muscles controlling the hand can give a misleading picture of the
degree to which the fingers can be controlled individually; however, there are several
factors that mitigate against this. One is that it is often not possible to control the
muscles that move the different fingers entirely independently. As has already been
mentioned, the muscle bellies of the deep and superficial flexors and the common
extensors of the fingers are partially fused, thus, contraction in any one of these is
likely to produce some passive movement in the others. The degree of fusion varies
between individuals and in the case of the flexors, slips of muscle may even cross
between the deep and superficial muscle masses. Another important restriction to the
independent control of finger movement is the degree to which the subdivisions of
the muscle can be controlled separately by the nervous system. Studies of the deep
flexors have revealed that in nonmusicians there is frequently a strong synchroniza-
tion between the activity of motor neurons controlling the parts of the muscle acting
on different fingers (Reilly, Nordstrom, and Schieber 2004). This is most noticeable
for the bellies supplying adjacent fingers and is particularly strong between those of
the ring and little fingers, whose tendons are also the most closely linked together.
The degree of motor neuron synchrony appears rather less for the subdivisions of the
common extensor of the fingers. You will be relieved to discover that studies of this
muscle indicate that with practice it is possible to reduce the level of synchronization
(Schmied, Ivarsson, and Fetz 1993; Schmied et al. 2000), which will allow the devel-
opment of greater finger control.
Connections between tendons running to different fingers are also a significant
feature of the hand. We have already encountered the extensive linkages between the
tendons of the common extensor muscle; however, atypical connections are common
and are rarely covered by standard anatomy textbooks. The common extensor tendon to
the ring finger may be connected to the tendon of the separate extensor of the little fin-
ger, thus restricting its independent action (Allieu et al. 1998). One study has suggested
that this may occur in up to 18 percent of individuals while in a further 34 percent, the
separate extensor muscle of the little finger is entirely absent (Baker et al. 1981). What
is less well appreciated is that within the carpal tunnel, extensive connections often ex-
ist between tendons of the deep flexor muscle running to the different fingers (Leijnse
et al. 1997). These can be either fine tendinous linkages or adherent sheets of tenosy-
novium. Though the degree to which this affects finger movement during playing has
62 Chapter Three

not been assessed, tensions put on these structures during playing could make them
potential sites of pain and inflammation. One study of music students revealed that the
divisions of the deep and superficial flexor muscles supplying different fingers often
show a considerable degree of dependence (Miller, Peck, and Watson 2002). Though
this rarely affects the index finger, its incidence progressively increases for the second
to fourth fingers, especially for the deep flexors. These were, however, asymptomatic
in terms of discomfort or pain. Some attempt has been made to model the effect of
such tendon linkages mathematically in the hope that it may ultimately be possible to
identify the types of finger exercise that are capable of improving performance (Leijnse
et al. 1992, 1993). Other exercises, which because of anatomical restrictions would be
incapable of improving dexterity (and which might even cause tissue damage), could
be discarded. For example, attempts to develop the ability to raise the fourth finger as
high as the others despite the restrictions imposed by tendon linkages may be a particu-
lar source of injury (Brown 2000). Techniques that optimize the influence of the small
intrinsic muscles of the hand (particularly the interosseus and lumbrical muscles), over
which greater individual control can be exerted, may be useful for pianists and have
been proposed by some teachers (Beauchamp 2003a).
Though the movements of the thumb are usually quite independent of those of the
fingers, it is not unusual to find an anomalous linkage between the tendon of the long
flexor of the thumb (flexor pollicis longus) and the deep flexor tendon of the index fin-
ger. This is known as Linburg-Comstock syndrome. Various studies have suggested
an incidence of 20–35 percent in the general population, with about a quarter of cases
showing the linkage in both hands. It is easy to demonstrate that this condition is quite
common: ask a group of musicians to curl up their thumbs and to tell you if the index
finger moves involuntarily (see video 3-1). Most affected individuals do not comment
on it, as they assume that it is the norm. The degree of linkage varies considerably,
and in a small proportion of instrumentalists it can lead to pain or difficulty in play-
ing (Miller et al. 2003). This is one of the few tendon linkages where surgery is both
feasible and generally beneficial (Allieu et al. 1998).
The considerable anatomical variation in the muscles and tendons in the hand is
therefore an important factor in performance that should be both recognized and ac-
cepted by musicians and their teachers (Bergman, Afifi, and Miyauchi 2004). Certain
muscles or tendons may be absent in some individuals, or their form or attachments
may differ from the norm. The superficial flexor digitorum tendon to the little finger
is missing in about 5 percent of hands (Miller et al. 2003). Variation is particularly
common in the lumbrical muscles, where as many as 50 percent of hands do not show
the standard pattern (Perkins and Hast 1993). The tendon of the third lumbrical muscle
divides to insert into the ring and middle fingers in up to a third of the population,
while in a few individuals there is no lumbrical insertion on the little finger at all. In
addition, the range of movement of different joints in the hand varies, as does overall
hand size. As a result, regardless of the degree of training, not all musicians are ca-
pable of moving the fingers in precisely the same way, and this should be reflected
in a flexible approach to the teaching of playing technique in general and keyboard
fingering in particular. Some practical examples of these problems and how they can
be overcome are discussed by Beauchamp (Beauchamp 2003b).
Figure 3-14. The paths of the radial, ulnar, and median nerves in the arm. These all arise from
the brachial plexus, whose roots emerge from between the cervical vertebrae. A. View from the
back. B. View from the front. The dashed circles show the most common sites of nerve compres-
sion or entrapment: the thoracic outlet where the nerves emerge from under the collarbone, the
ulnar nerve at the elbow, and at the wrist, the median and, to a lesser extent, the ulnar nerves.
64 Chapter Three

The Nerve Supply to the Arm and Hand

The nerves running to the arm originate from the cervical (neck) region of the spinal
cord. They emerge from between the lower neck vertebrae as five nerve roots, which
fuse and divide in a complex manner to form the nerves supplying the arm and shoul-
der. This cluster of nerve roots and their branches is called the brachial plexus. We
will consider only the three largest nerves that emerge to supply the arm: the radial,
medial, and ulnar nerves (fig. 3-14). To reach the arm, the nerves must pass through
a narrow gap rather inappropriately called the thoracic outlet, which lies between the
collarbone and the uppermost rib. As well as the nerves, this crowded space contains
the main artery and vein supplying the arm and gives its name to a number of condi-
tions (thoracic outlet syndromes) that can arise if either the nerves or the blood vessels
become compressed here (see below).
The radial nerve spirals round the back of the upper arm, through the triceps mus-
cle, which it supplies. When it emerges on the lateral side of the elbow, it crosses in
front of the joint before turning to the back of the forearm to supply muscles there. It

Figure 3-15. The distribution of the ulnar and median nerve branches in the hand. The ulnar
nerve supplies the muscles at the base of the little finger, the interosseus muscles, two of the lum-
bricals, and the adductor of the thumb. The median nerve supplies the muscles at the base of the
thumb and the remaining lumbricals. The arrows indicate where the nerves may become trapped:
Guyon’s canal, a passage through the thickness of the retinaculum, for the ulnar nerve and the
carpal tunnel for the median nerve. Nerve compression may produce tingling or numbness in the
areas of skin supplied, indicated by the shaded regions in the small hand outlines.
 The Shoulder, Arm, and Hand 65

therefore controls the extensor muscles of the elbow (triceps), wrist, and hand, which
straighten the elbow and fingers and pull the wrist back. Of the three large nerves of
the arm, the radial nerve is the least often damaged in musicians. However, because it
wraps tightly round the humerus, it may be damaged if the bone is fractured.
The median nerve supplies mainly flexor muscles. It passes down the front of the
upper arm underneath the biceps (though it doesn’t supply it). Biceps and some asso-
ciated muscles that flex the upper arm and elbow are controlled by the much smaller
musculocutaneous nerve. The median nerve crosses in front of the elbow and between
two parts of a muscle that pronates the forearm (pronator teres). This is an important
relationship, as the median nerve can sometimes be entrapped and compressed by the
muscle. The nerve branches extensively in the front of the forearm to supply flexor
muscles of the wrist, fingers, and thumb. It enters the hand by passing under the reti-
naculum at the wrist (fig. 3-15). This narrow channel (the carpal tunnel) is one of the
most common sites of median nerve compression. Within the hand, the median nerve
supplies the pad of short muscles that move the thumb (the thenar eminence) and the
lumbricals of the index and middle finger. It is also sensory to the skin on these two
fingers, the thumb, and the side of the palm from which they arise. Compression of
the nerve in the carpal tunnel may therefore produce a tingling or numbness in this
area of skin and weakness or wasting of the muscles it supplies. The middle of the
palm remains unaffected because the branch of the median nerve that provides it with
sensation runs over the top of the retinaculum.
The ulnar nerve runs down the inside of the upper arm but supplies no muscles
there. At the elbow, it passes behind the broad part of the humerus on the inner side
just above the joint before passing into the front of the arm in a channel (the cubital
tunnel) between the two heads of one of the wrist flexors (flexor carpi ulnaris). The
ulnar nerve is very vulnerable here (as most of us know to our cost). If the elbow is
struck forcibly against some object, the nerve may be pinched against the bone, which
leads to pain and a tingling sensation in the little finger and the adjacent side of the
ring finger, the region of the hand whose sensation is carried by this nerve. It can also
become entrapped in the cubital tunnel (see animation 3-6). In the forearm it supplies
one of the wrist flexors and contributes to the innervation of the deep flexor muscles
of the fingers. At the wrist it passes under part of the retinaculum (Guyon’s canal),
where it is again vulnerable to compression, then supplies most of the small muscles
of the hand (the short muscles to the little finger, all the interosseus muscles, the lum-
bricals for the fourth and little fingers, and the adductor of the thumb) (fig. 3-15). If
the nerve is badly damaged at the elbow or in the hand, the resulting weakness in the
lumbricals and interossei makes it difficult or impossible to straighten the little and
fourth fingers, which curl into the palm as a result.

PLAYING-RELATED INJURY—YOU ARE NOT ALONE

The idea that certain precisely definable injuries are associated with particular profes-
sions is not a new one. For example, the Mad Hatter in Alice in Wonderland exhibits
the symptoms of poisoning by the mercury compounds that were used in preparing
66 Chapter Three

fur for the cheaper varieties of top hats. Musculoskeletal problems associated with the
maintenance of awkward postures or with repetitive movements were first recognized
in a remarkable and wide-ranging study of occupational diseases by the seventeenth-
century physician Bernardini Ramazzini (Franco 1999; Ramazzini 1964), though his
findings were perhaps not generally appreciated by the medical community at that time,
and certainly not widely acted upon for a further three hundred years. Over the last
twenty to thirty years, there has been a growing awareness among orchestral managers
and among some medical professionals that musicians are particularly vulnerable to
performance-related injury. This awareness has arisen from the work of a few musi-
cians who have risked their careers by publicizing their playing-related problems and
by seeking to devise and implement practices in their own orchestras aimed at reduc-
ing the incidence of injury and promoting successful rehabilitation of injured players
(Paull and Harrison 1997; Culf 1998; Horvath 2002, 2004). Awareness of these issues
among players and managers has been aided by the emergence of arts medicine as a
distinct discipline and the recognition that the treatment of musicians can be a special-
ized area for clinicians (Winspur and Wynn Parry 1998; Sataloff, Brandfonbrener, and
Lederman 1998) in the way that sports medicine has long been. Indeed, the injuries
experienced by musicians, as well as their treatments and prevention, have much in
common with those of athletes. In the UK, schemes such as the Healthy Orchestra
charter, developed by the Association of British Orchestras, the Musicians Benevolent
Fund, and the British Association of Performing Arts Medicine, are being developed
to promote musicians’ health.
One of the seminal events that awakened interest in the medical problems of musi-
cians was the publication in 1988 of the International Convention of Symphony and
Orchestral Musicians Survey, which was based on questionnaires sent to over four
thousand musicians (Fishbein et al. 1998). This was published in the newly formed
journal Medical Problems in Performing Artists. There is, of course, always a danger
with such surveys that the responders are a self-selecting population with injury prob-
lems that may not be typical of the whole population contacted. Also, many groups
of musicians such as keyboard players, singers, and jazz or rock performers would
have been underrepresented. Nevertheless, more than 55 percent of those contacted
replied, which suggested that health issues were a major concern for these musicians.
According to Wynn Parry (1998c), quoting summary data not in the original paper, the
survey revealed musculoskeletal problems of the neck, shoulder, arm, or hand in over
66 percent of string players, 48 percent of woodwind players, and 32 percent of brass
players. About half of these problems were rated “severe” by the players concerned.
A large number of subsequent surveys of professional musicians and both school-age
and university-level music students found incidences of performance-related pain or
musculoskeletal problems of 34 to 87 percent, though in most cases severity was not
assessed (Zaza 1998). Perhaps even more telling is a report of the experiences of a
group of amateur musicians who went on a one-week performance course. The play-
ers who normally practiced only an hour a day played for six to seven hours daily
for the week, and as a result 72 percent developed new and painful playing-related
symptoms (Newmark and Lederman 1987). Although this is an isolated example, it is
indicative of the consequences that can be expected from a rapid scaling up of playing
 The Shoulder, Arm, and Hand 67

intensity, a pattern of behavior that is likely to be common in musicians preparing for

an imminent recital or audition (Newmark and Hochberg 1987) or among students on
first entering the competitive environment of a music school. Nevertheless, most of
the plethora of surveys of musicians’ injuries rely on self-reporting of symptoms by
the participants and therefore lack any objective medical assessment of the physical
problems reported or, indeed, comparison with an age-matched control population of
nonmusicians. They therefore take us little further in understanding the actual range
and severity of playing-related injury.
Many musicians find it possible to continue to play in the face of upper-limb pain
and may even accept it as an inevitable part of their profession. However heroic this
may seem, it is ultimately counterproductive as, without diagnosis and treatment, the
condition is likely to deteriorate until there is a clear loss of technique through tis-
sue damage, muscle weakness, or an attrition of fine motor control (Fry and Rowley
1992). We have already seen in chapter 1 that pain is usually linked to inflammation,
which exacerbates many of the injuries that musicians are likely to experience. In ad-
dition, changes in posture to compensate for loss of dexterity or restriction of move-
ment frequently lead to difficulties at sites other than the original source of discomfort.
This not only compounds the problem, but can make it much harder to diagnose the
root cause. Symptoms such as extensive pain, loss of dexterity in the playing move-
ments, tingling or numbness, muscle weakness, or any indication of changes in the
shape or bulk of muscles in the arm or hand must be taken very seriously and help
sought as rapidly as possible.
There may be many reasons why players stoically soldier on in the face of pain.
First, in a few dark corners, the outmoded concept of “no pain—no gain” may still
hold sway. The overview of pain in chapter 1 should have amply demonstrated that
this is an inappropriate response. It certainly isn’t applied to professional athletes,
who are valuable commodities financially and whose teams therefore demand the
most rapid and effective treatment for their injuries. Discomfort from exhaustion
is one thing (and is probably what the old adage originally referred to), but pain
as a result of tissue damage is quite another. Other reasons why musicians try to
ignore pain may be a fear of revealing a physical weakness in a highly competitive
profession, or a worry that they may be ordered to cease playing for a prolonged
period or be encouraged to agree to a risky intervention such as surgery. Clinicians
who are familiar with musicians’ problems will be well aware of the psychological
consequences of an embargo on playing. As to surgery, this is always a last resort
and is appropriate in only a small proportion of cases. Treatment is more likely to
involve a combination of less daunting therapies, and a central plank of this should
always be reassessment of playing posture and technique, as well as a consider-
ation of general health and fitness and an exploration of strategies to relieve stress,
which is a major contributing factor to playing-related pain. Some of the treatment
may even be directed at the instrument, to adapt it more closely to the body, or the
physical limits of the player. Figure 3-16 provides a summary of the range and pos-
sible interactions of factors underlying performance-related problems experienced
by a group of musicians who sought help at one well-known clinic specializing in
performance-related problems.
68 Chapter Three

Figure 3-16. An analysis of the causes of playing-related symptoms and treatment of patients
seen at a clinic run by Winspur and Wynn Parry (reproduced with permission from Winspur and
Wynn Parry [1998], p. 38, fig. 2).

SEEKING HELP WHEN THINGS GO WRONG

Perhaps the most vital message to take on board about seeking help is to seek it early.
The more established an injury is, the harder it will be to diagnose and treat, and the
longer and more difficult will be the rehabilitation. It is also often important to seek
out a clinician who is not only a specialist in the relevant medical discipline (e.g,. an
orthopedic or neurological consultant), but also is familiar with these problems as they
are manifested in the musician. The role of the general practitioner is to treat com-
mon ailments and to identify those that need the attention of a specialist, to whom the
patient will then be referred. In most instances, the general practitioner will not be
familiar with the specialist requirements of musicians, and it is unrealistic to expect
this. In turn, hospital specialists, though undoubtedly possessing a detailed knowledge
of a particular field of medicine, may see few musicians among their patients and may
be more attuned to occupationally related limb problems among manual laborers or
the elderly, if these make up the majority of their patients. A specialist will not ask a
bricklayer with a suspected rotator cuff tear to bring in his hod, but someone with a
knowledge of the upper-limb problems of musicians will certainly need to examine
the patient with his or her instrument in order to understand the posture that player
adopts and whether this needs correcting. If your doctor does not want to see your
playing posture, you may wish to consult elsewhere. It is also vital that a detailed
patient history is taken that takes into consideration all aspects of a musician’s life-
style. In a retrospective analysis of a group of over six hundred musicians presenting
with upper-limb problems at a clinic specializing in musicians’ problems, only 41
 The Shoulder, Arm, and Hand 69

percent turned out to have a diagnosable physical condition (fig. 3-16). In 40 percent
the symptoms arose from problems with technique, and in 19 percent the physical
symptoms appeared to be a manifestation of stress and anxiety (Winspur and Wynn
Parry 1998).
The specialist clinician will also apply a knowledge of the patient’s instrument
and lifestyle to devise a suitable treatment. This will take into consideration the im-
portance of keeping any reduction in playing to the minimum that is compatible with
a successful outcome. A well-planned rehabilitation program is not only important
for therapeutic goals such as maintaining the mobility of affected joints, but also for
retaining technical skills and ensuring the mental well-being of the patient. In addi-
tion, where surgical intervention is necessary, it is important that the way in which
this is carried out has as little impact as possible on playing, and this may necessi-
tate a rather different approach to that used for the same condition in a nonmusician
(Winspur and Wynn Parry 1997). As will be seen from some of the later sections of
this chapter, diagnosis of neuromuscular problems in the upper limb can be difficult
and the symptoms are sometimes confusing, so unraveling the primary cause of the
malady may require considerable insight. Furthermore, though the conditions seen
in musicians occur quite frequently among nonmusicians, the symptoms may not be
identical (Brandfonbrener 2003). Fortunately, clinicians with the necessary skills are
increasingly to be found as arts medicine becomes an established discipline. The field
is supported by a growing body of knowledge of the health problems of musicians that
has come from research in a number of specialist centers that have arisen in Europe
and the United States.
Fully resolving a playing-related problem is likely to involve a number of different
therapeutic strategies. Your doctor will start the ball rolling by assessing and treating
your immediate physical symptoms. This will often be followed by physiotherapy
over a period of weeks or months to strengthen the muscles that give postural sup-
port to a joint that is causing problems, or to relieve pressure on a compressed nerve.
Some of these exercises may be continued indefinitely to maintain muscle strength
or joint flexibility. However, the primary cause also needs to be established, together
with any factors that may have aggravated the original injury. If this is not done, it
is likely to occur again, or at the very least, apprehension that it could limit or affect
playing will remain. You may want to be a little less enthusiastic with your home-
improvement or holiday pursuits if you think that these have been triggering factors,
but even where there is an obvious starting point, you should reexamine your tech-
nique with your instrumental teacher or mentor and with a physiotherapist. This may
result in some retraining during which a temporary reduction in the fluency of play-
ing must be accepted for long-term gain. Remember that playing with good technique
should ultimately improve performance! You may find it beneficial to use strategies
such as the Alexander technique, Feldenkrais method, yoga, or tai chi to promote
greater postural awareness and reduce stress, which is beneficial both for general
emotional and physical well-being and for reducing chronic muscle tension.
Information on where to find doctors, physiotherapists, or dentists who specialize
in the problems of musicians can be found in a number of places. In the UK, help can
be obtained from the British Association for Performing Arts Medicine (BAPAM,
70 Chapter Three

Totara Park House, Fourth Floor, 34-36 Gray’s Inn Road, London WC1X 8HR; Tel.:
0845 602 0235, website: www.bapam.org). The equivalent organization in the United
States is the Performing Arts Medical Association (Mary Fletcher, Executive Direc-
tor, PAMA, P.O. Box 61228, Denver, CO 80206; Tel./Fax: 303-632-9255, e-mail:
artsmed@comcast.net, website: www.artsmed.org). Many other useful sources can be
found in the extensive resource list in Horvath (2002).
The details of treatment for individual conditions will vary according to your
doctor’s assessment of the symptoms and the precise diagnosis made, but the basic
palette of treatments is explained here, though the list is far from exhaustive. This
is done partly to avoid repetition in the following sections on specific ailments, but
mainly to provide a general overview of the short-term consequences of seeking help.
The object is to try to dispel unnecessary fears of how drastic treatment may be and the
extent to which it may disrupt musical life, while still giving a realistic picture of what
is likely to be involved. It is hoped that by clarifying these issues, players will be less
resistant to seeking help in a timely fashion. The way these therapies are now applied
owes a considerable amount to the development of sports medicine, which has played
an important role in advancing the treatment of musculoskeletal and related problems.
It is not only important to get athletes rehabilitated as rapidly as possible, but also to
prevent a rapid recurrence of the injury. The same principles apply to the rehabilitation
of musicians, even if they cannot command the same transfer fees!

Rest
Apart from the prospect of surgery, being advised to abstain from playing for a pro-
longed period of time may be the form of treatment most dreaded by musicians, for
whom this may at best result in loss of facility through lack of practice, and at worst,
loss of identity, self-esteem, and the ability to earn a livelihood. At one time, rest
was the most commonly prescribed treatment for a wide range of musculoskeletal
problems. The joint, tendon, or muscle was simply immobilized as much as possible
until the condition resolved. While it remains true that for many conditions we still
rely on the body to repair itself by its own internal mechanisms, rest is no longer seen
as a panacea, and in most cases muscle wastage (which happens very rapidly), joint
stiffness, and other consequences of immobility are positively disadvantageous for
optimum recovery. Furthermore, after surgery to free tendons and nerves, it is particu-
larly vital that physiotherapy and the control of inflammation start as soon as possible
to prevent scar tissue from binding together structures that need to move relative to
one another. The initial stages of treatment are aimed at reducing the acute symptoms
(i.e., immediate symptoms of pain and inflammation that arise abruptly and may abate
quite quickly). Nowadays, experienced practitioners dealing with musicians do not of-
ten advocate prolonged periods of complete rest; however, it is vital that the return to
playing is carefully managed and that agreed timetables are strictly adhered to (Win-
spur 1998d). A musician with a serious problem may start playing again for as little as
five minutes twice a day. Progress is carefully monitored and playing time increased
accordingly. Following such a timetable may be frustrating initially, but it should be
remembered that this will ultimately provide the quickest route to recovery. Though
 The Shoulder, Arm, and Hand 71

the periods of actual playing may be restricted, the player will benefit both musically
and psychologically from perfecting the skills of mental rehearsal (see chapter 8) and
in learning and analyzing scores that will be played on recovery. Of course, an even
better approach is not to delay seeking help until the injury has developed to the extent
that this is required. Practical advice for dealing with the period of rehabilitation can
be found from a number of sources (Horvath 2002; Warrington 2003). The importance
of a staged rehabilitation from injury is now recognized by some orchestras, which
have put in place procedures that allow a progressive return to full playing (Horvath
2004). These are the orchestras that, even in a difficult financial climate, truly value
their players and take a rational view of injury and its effect on performance. It is to
be hoped that such practices will in time become the norm, as they are more likely to
result in a full recovery and prevent an avoidable loss of career in injured musicians.

Treatment for Acute Pain and Inflammation

The acute symptoms of injury are often pain and swelling of the affected area. These
two symptoms are related to a considerable degree, because the nerve endings that
carry information on pain also release chemicals at the injury site that contribute to the
inflammation (see chapter 1). Any restricted spaces within the tissue through which
structures such as tendons or nerves must slide will become narrower, so that friction
is generated by such movement. This creates a vicious circle in which further damage
leads to more pain and inflammation. Rapid control of inflammation is therefore im-
portant. One anti-inflammatory treatment you will all have seen on the athletic fields is
the application of cold packs. This constricts the blood vessels that are the source of the
accumulating fluid, slowing the development of the inflammation. It is most effective
when applied as quickly as possible after the injury or immediately when the swelling
first appears. That is why injured athletes will apply an cold pack even when watching
the remainder of the game in which they have been injured. A bag of frozen peas makes
a good substitute for home use, as it can be molded to the shape of the injured region;
however, if it is very cold, it may be necessary to put a cloth between the pack and the
skin to prevent tissue damage. A typical strategy would be to apply the pack for twenty
to thirty minutes every three to four hours. Inflammation can also be reduced by raising
the swollen area to reduce the blood pressure that is driving the fluid into the damaged
tissues. For the arm and hand, it is particularly effective if these can be raised above the
head. Compression of the swollen region with elastic bandages will also help to prevent
the fluid from accumulating. Oral drugs with both painkilling and anti-inflammatory
properties (e.g., ibuprofen) may also be prescribed, while in more severe chronic condi-
tions, such drugs may also be injected into the injury site by a skilled clinician.

Splinting
During rehabilitation from injury, splints may be employed to provide support or to
transfer load from one part of the limb to another (see fig. 3-17). For example, in a
clarinetist, a splint can be used to transfer the weight of the instrument from the last
joint of the right thumb, where it is usually carried, to the base of the thumb, where it
72 Chapter Three

Figure 3-17. Different types of splints used on the fingers. A. Splint made from heat-molded
plastic, which allows molding of the shape. This one supports the joint at the base of the thumb.
B. A dynamic splint made from a flexible adhesive dressing, used to limit the degree of movement
of the metacarpophalangeal joint. C. A ring splint made of plastic or metal used to support the
finger joint at a fixed angle. This is useful where joint hypermobility might cause the finger joint
to collapse under the loading experienced during playing.

exerts less leverage. Splints can also be used to keep an injured joint straight if there is
a tendency for the damaged tissues to contract, which leads to chronic flexion. For the
fingers in particular, the use of splints can greatly reduce the time needed to regain full
extension of a joint after injury. Nowadays, rigid splints can take a variety of forms,
some of which are shown in figure 3-17. Many are made of a stiff but thin thermo-
plastic, which when warmed can be molded to the precise shape required. This is very
useful for producing small, irregularly shaped splints to support the fingers, and also
ensures a comfortable fit. Where it is necessary to fix or support a finger joint during
routine playing as opposed to during rehabilitation or reeducation, ring splints can be
worn. These are sets of metal or plastic rings, similar in thickness to thin jewelry rings,
that are welded together so that when slipped onto the finger they lie round the bones
on either side of a joint, holding it at a fixed angle. Dynamic splints (which serve to
limit the range of movement of a joint) can be fashioned from adhesive tape to fix the
end point of joint movement to help players learn new playing practices or to avoid a
joint angle that may be risky (Stotko 1998).

Physiotherapy
Physiotherapy plays an important role in strengthening muscles and increasing joint
flexibility, either to address directly the causes of a playing-related problem or as part
 The Shoulder, Arm, and Hand 73

of the process of rehabilitation after surgical or nonsurgical treatments. The muscles

that are used in playing will be well developed but others that are important for posture
may be less so and may need strengthening to increase hand or arm stability. When
movement becomes restricted due to problems with tendons or joints, or because of
nerve damage, muscles rapidly get smaller and weaker. Rebuilding them through a
program of exercise is a much slower process, and it is important that this is carried
out conscientiously. Similarly, regaining the full mobility of a joint that has been
swollen or a tendon that has become tight in its sheath can also be difficult and some-
times painful, and a steady approach through physiotherapy is the only recourse. The
boundary between physiotherapy, exercise, and body maintenance is a blurred one;
activities such as the Alexander technique, Feldenkrais, yoga, and tai chi may start
off in one role and end up in another. Equally, any aerobic exercise such as walking,
jogging, or swimming can be viewed as physiotherapy, as these activities move joints
and muscles in a different way from that required by playing an instrument. These
forms of exercise will ultimately benefit your musical activities, not only through their
specific effects on particular muscles and joints but also by improving general health
through their conditioning effects on the cardiovascular and respiratory systems.

Surgery
For most conditions, surgery is the last resort, and it is important that before this course
is decided upon, a most careful diagnosis is carried out to ensure that it is the most
appropriate course of action. Under such review, surgical treatment is uncommon. In
the above-mentioned sample of over six hundred musicians with upper limb problems,
only 4 percent underwent surgery (see fig. 3-16) (Winspur and Wynn Parry 1998). As
already mentioned, the surgical approach used for a particular condition in a musician
may differ from that in a nonmusician. One important consideration is the site of the
incisions that are to be made through the skin (Winspur 1998d). For example, these
should not cross areas of skin used to close open holes on wind instruments or to hold
down strings. Here the nature of the player’s instrument must be considered carefully.
For example, while in most wind instruments it is the pad of the last joint of the finger
that is used to close the hole, in the bagpipe the skin on the middle joint is used in all
but one finger. Similarly, incisions in any part of the finger or thumb that supports the
weight of the instrument should be avoided if possible. It is equally important not to
damage small nerve branches that carry sensory information from areas of the skin
that contact the instrument.
The advantages of undergoing surgery under local anesthesia are increasingly fa-
miliar to us today. Patients are usually discharged on the same day as the operation,
and the small degree of risk associated with general anesthesia is eliminated. In the
case of operations on the musician’s hand this approach has additional psychologi-
cal advantages; for example, the patient can be kept informed and reassured as to the
progress being made throughout the operation. To some extent the result of the in-
tervention on active finger movement can also be checked before wound closure, if
this is appropriate. All this gives the patient a greater feeling of control over what
is happening, which can be very reassuring for those whose livelihood depends on
74 Chapter Three

a high level of hand function. Another very important consequence of hand surgery
under local anesthesia is that the patient can, by placing the hand on the head, keep
it well elevated from the instant the operation is completed. Though this may sound
rather bizarre, it substantially reduces the degree of swelling and consequent pain that
is experienced postoperatively, which considerably speeds recovery (Winspur 1998d).
Unless the operation has been to repair traumatic accidental damage to tendons, it may
be possible to start playing for a few minutes several times a day only four or five days
after the operation. Musicians faced with the possibility of surgery of the arm or hand
would do well to read a short paper by Winspur and Wynn Parry (2000) before as-
sessing their options. The clinician should provide a realistic assessment of the risks as
well as the benefits of surgery in the context of the high level of functionality required
by today’s professional musician. This will be very different from what is likely to be
acceptable to most nonmusicians undergoing a similar treatment. Informed consent
must be just that.

WHAT CAN GO WRONG?

Generalized Pain and Overuse Syndrome

One of the most common diagnoses given for pain among musicians (and indeed those
in other professions requiring rapid repetitive movements over prolonged periods)
is overuse syndrome. This term, which is commonly and perhaps sometimes glibly
used, may often not really reflect the root of the problem. As a diagnosis it has been
the subject of a considerable amount of debate among clinicians, which has tended
to generate more heat than light. The usual definition given is of a condition arising
as consequence of exceeding the biological or physical limits of the tissue (e.g., of
a muscle, tendon, nerve, or joint). This doom-laden description suggests imminent
physical collapse and the end of a career, but increasingly it appears that the defini-
tion probably misses the point. It is possible to play for years without problems and
then start to experience pain. This does not generally mean that a body component
has suddenly worn out; however, an injury to a tendon in the wrist, for example, no
matter how it is acquired, is likely to be exacerbated by the constant rapid movement
of the fingers that is the normal daily life of a musician. Perhaps “delayed treatment
syndrome” might be a more illuminating term in many cases, as prompt recognition
and diagnosis of the injury will go a long way to prevent the appearance of chronic
symptoms. Overuse syndrome has sometimes been linked to repetitive strain injury
(a term that will probably also be familiar to the reader), whose diagnosis at one time
threatened to reach epidemic proportions. This name is now little used by clinicians,
as it has become apparent that symptoms formerly associated with this condition are
not necessarily caused by repetitive movements in themselves (for the reasons just
given), nor by any form of strain (Wynn Parry 1998b).
The demands on the body experienced by musicians have many similarities with
those of athletes, so let’s illustrate this with a sporting analogy. Let us suppose that
after keeping fit for years by running, you decide you want to get your exercise by
 The Shoulder, Arm, and Hand 75

rowing instead. You regard yourself as fit, so when you get in the boat or rowing ma-
chine, you are always pushing to achieve the same level of exercise you used to get
from your daily five-mile run. Your performance in the boat improves in a gratifying
manner over the first few sessions, then suddenly you start to get low back pain and
feel discomfort in the middle of the back and one side of the chest. You put this down
to the usual muscle discomfort associated with getting fit, but after a couple more
sessions you have to admit that you have pulled a muscle connecting the shoulder
blade to the ribs and must stop rowing. This and the back pain are technically overuse
injuries, but only in the sense that you have exceeded the capabilities of the muscles
at their current level of training. You don’t imagine your sporting or even rowing
days are over. There is no mystery about the source of the problem or its remedy. You
realize that though your leg muscles are strong from years of pounding the pavement,
those of your trunk and upper body are not at the same level. You let the injury heal,
get some advice on technique, and return to rowing through a carefully monitored
series of sessions, gradually increasing their length and the effort they require. This is
not only for the benefit of the injured muscle, but also to ensure that your other upper-
body muscles are not overstressed as they gradually increase in strength. In addition,
you make sure you warm up thoroughly before you row strenuously. In this context
you will recognize that the approach is simple common sense.
Now let’s imagine a different scenario. You are a young musician and you have
just entered a prestigious music school. You thought you were a pretty good player
and are shocked to discover the standard of “the competition.” Overnight you double
your daily practice. Your arms are stiff when you wake up in the morning, and instead
of improving, your playing gets worse because your fingers don’t seem to want to
respond the way you want them to. Not wishing to be defeated, you redouble your
efforts and start to experience pain in the forearm and wrist, which finally makes play-
ing almost impossible. The doctor gives the dreaded diagnosis of overuse syndrome
(or perhaps some other condition with an even more impenetrable name). This no
more means that your body can’t cope with the life of a professional musician than
was the case with the athlete. The remedy is likely to be similar to that of the previous
example, but how much better it would have been to have avoided the breakdown by
opting for a more measured approach to increasing practice. It is easy to say and hard
to do, perhaps, but returning to full playing after injury takes time and may sap your
confidence, so a carefully staged increase in daily practice will achieve your objec-
tives more rapidly in the long run. From a psychological point of view, a good way to
approach improving your level of skill in this situation is to set out a timetable with a
regular series of goals. You can then pick them off one by one, and have confidence
that after a given period of months you will have a good chance of achieving the level
of performance you have set for yourself.
While there are many specific problems that arise in players of instruments that
require rapid and sometimes forceful finger movements over prolonged periods of
time, the underlying physical changes that result in overuse syndrome remain obscure.
This situation is not aided by the fact that it has no widely agreed definition (Fry and
Rowley 1992; Rozmaryn 1993; Brandfonbrener 2003; Winspur 2003). This is not
a trivial problem, because a condition described only on the basis of a set of rather
76 Chapter Three

generalized symptoms may in fact represent a number of distinct conditions with dif-
ferent origins, each requiring a specific treatment. It is not even clear that it is actually
the intensity of playing that is the problem, rather than the abrupt changes in patterns
of practice. One’s susceptibility to it may be strongly influenced by overall fitness,
stress levels, and other lifestyle issues. A major contributory factor is likely to be pos-
ture. The one adopted, which should be tailored to your particular body shape, should
minimize the chance of physical problems developing. Good posture in this sense will
naturally form the basis of an efficient playing technique. A maintained high intensity
of playing need not result in overuse syndrome as long as this is reached by gradual
increments and if the technique is good, the lifestyle healthy and stable, and sensible
practice and playing patterns including warm-up, stretching exercises, and periodic
rest periods are adhered to (Horvath 2002; Lippman 1991). For a short, thoughtful
account on overuse syndrome, the reader is referred to an excellent article by Wynn
Parry (1998b).
The pain associated with overuse syndrome may initially be felt in the hand during
playing but may later spread to the forearm and shoulder. One characteristic feature
of this pain is that it is not restricted to any particular joint, tendon, or muscle group.
As the condition deteriorates, pain may be experienced during movements involved
in everyday tasks such as turning a door handle. Strangely, though the pain can be
severe, its physical characteristics are often hard to pin down. For example, other than
some slight muscle tenderness, there is usually no real muscle weakness or restriction
of movement, and there may be no indication of tendonitis or loss of sensation. The
condition has variously been attributed to biochemical changes in the muscle result-
ing from extreme fatigue (Wynn Parry 1998b), the breakdown products released from
damaged, swollen muscles (Fry 2000), damage to the junction between the muscle and
tendon (Lederman and Calabrese 1986), or inflammation of the synovial sheath of the
tendon (Hochberg et al. 1983), which is actually tenosynovitis (see below). None of
this confusion, however, detracts from the very real pain and disability that is experi-
enced. Furthermore, chronic pain can cause long-term changes in the circuitry of the
spinal cord, which may result in sensations that were previously innocuous becoming
painful (chapter 1). This adds to the difficulties of resolving the problem.

Problems with Tendons

One specific problem that is sometimes linked to overuse is tendonitis. Though ten-
donitis literally means “inflammation of the tendon,” in practice the symptoms may
arise from inflammation of the synovial sheath of the tendon (the tenosynovium),
in which case it should be known as tenosynovitis, or degenerative changes in the
tendon even in the absence of inflammation. To make matters more confusing, many
additional names are used to describe tendonitis of different types at particular sites in
the upper limb (Rozmaryn 1993); as these are commonly in use by clinicians, several
of the names will be employed in the following account. As with many apparently
common conditions, vigorous debate continues as to the origin of the several forms
of tendonitis. For example, though many consider that tendonitis can be a manifesta-
tion of repetitive strain injury, there is still a lack of objective evidence to support this
 The Shoulder, Arm, and Hand 77

(Wynn Parry 1998e). Among musicians, tendonitis is not as common as one might
expect given the repetitive nature of their playing movements, and when encountered
in this group it often appears to have been triggered by a nonmusical activity (e.g.,
home-improvement project or sports), perhaps because these may involve short pe-
riods of intense physical effort of a type that the body is unused to. For example, it
might be the consequence of a day spent putting up shelves or curtain track using a
manual screwdriver, which involves maintaining a strong grip on the screwdriver
coupled with repeated, forceful supination of the forearm against the resistance of
the screw. Such activity might cause tendonitis in the shoulder, elbow, or wrist. Paint
stripping and prolonged periods of vigorous sanding are also notorious for inducing
tendonitis (Winspur and Wynn Parry 1998). Once it has developed, tendonitis is likely
to be aggravated by the movements required for playing, though this in itself is not
the primary cause. Nevertheless, prolonged practice of technically demanding pieces
requiring rapid fingerwork can sometimes result in swelling and tenderness in the arm
or hand of keyboard and string players. This may involve not only the tendon and
tendon sheaths of the long flexor muscles of the fingers whose muscle bellies lie in
the forearm, but also (and this is sometimes overlooked) those of the small intrinsic
muscles of the hand, for example, the lumbrical and interosseus muscles (Silver and
Rozmaryn 1998).
Problems with the shoulder are frequently seen in players of bowed string instru-
ments because of the maintained elevated posture of the arm. If you watch the string
section of almost any orchestra you will see that the frequency and degree of right-arm
abduction in violinists and viola players, and of either arm in cellists, varies greatly
between individuals. Discomfort may be experienced in the tendons of the rotator cuff
muscles (rotator cuff syndrome), which connect the arm to the shoulder blade. This
often originates from what is known as an impingement, in which the tendon becomes
pinched against the bone. A common site of impingement is where the tendon of the
supraspinatus muscle passes through the narrow channel under the acromium of the
shoulder blade (animation 3-2). This may cause the tendon to swell due to inflamma-
tion, which only makes the impingement worse. The friction this generates can abrade
and ultimately tear the tendon through completely. This is most likely to occur in
people who are older or physically inactive, as muscles and tendons that are not used
quickly become reduced in bulk and strength. Among musicians, rotator cuff prob-
lems are not only confined to players, but may also affect conductors. The chance of
developing this and many other injuries is therefore reduced not only by good posture,
but also by maintaining a reasonable level of general fitness, which helps maintain the
muscles’ integrity. This is important not only for muscles used directly in playing but
also those that have a role in stabilization and support.
The extensor muscles of the fingers arise from a common flattened tendinous
structure that attaches them to a lateral ridge on the lower end of the humerus, just
above the elbow joint. This region is called the lateral epicondyle, meaning that it lies
above (epi-) the articulation of the joint (condyle). Pain symptomatic of tendonitis in
this region may occur after repeated forceful rotation of the hand and wrist, as in the
example given above of using a screwdriver, or it may be associated with extension
of the wrist while forcibly gripping the handle of something like a tennis racket or
78 Chapter Three

shovel. Though technically termed lateral epicondylitis, it is known more familiarly

as tennis elbow.
Another common site of tendon pain is at the base of the thumb, where the ten-
dons of the long abductor and extensor muscles of the thumb run in the walls of the
anatomical snuff box. Tenderness in this regions may be a sign of de Quervain’s
syndrome, an inflammation of synovial sheaths of the tendons. It is sometimes seen
in young female bassoonists and may also be present in other woodwind players who
must support the weight of the instrument on the thumb (most commonly clarinetists).
In keyboard players similar problems can be experienced in the flexor tendons of the
fingers of either hand where they cross the wrist. In violinists and guitarists it is ten-
dons of the left hand that are most commonly affected, particularly those of the middle
and ring fingers (Winspur and Wynn Parry 1998). This may be accompanied by some
transient numbness in the nearby skin. The tendon itself sometimes develops a nodule,
which makes it difficult for the tendon to slide through the synovial sheath as the fin-
ger straightens (fig. 3-13). Nodules on tendons also arise due to rheumatoid arthritis
or as a result of diabetes. When they occur in a flexor tendon, the finger tends to be
held crooked as if round a trigger, which is why it is given the name “trigger finger.”
When a persistent attempt is made to extend the finger, the nodule presses against the
entrance of the sheath until it finally forces its way in with a snapping sound.
Given time and a respite from inflammation, tendonitis will usually resolve itself.
The symptoms can also be treated effectively by injections of anti-inflammatory
drugs and local anesthetic into the tendon sheath, which usually provide dramatic and
instantaneous relief. Such treatment is generally followed up by physiotherapy in the
form of stretching and strengthening exercises (Winspur and Wynn Parry 1998; Silver
and Rozmaryn 1998). Ring splints may also be utilized to restrict tendon movement,
thus reducing the aggravation caused by the tendon continually passing through the
trigger point. Though surgery is sometimes advocated to open the tendon sheath, this
can carry particular risks for musicians. For example, in violinists, cutting open the
sheath to provide relief for de Quervain’s syndrome may leave the tendon prone to
dislocation from its channel because of the flexed and radially deviated wrist posture
that is sometimes required in either hand during playing (Winspur and Wynn Parry
1998).

Ganglion Cysts
Swellings under the skin of the wrist, called ganglion cysts, can cause inflammation
of tendons. The cysts, which usually contain a mucuslike fluid, most often occur on
the dorsal surface for the wrist. They are benign soft tissue tumors that sometimes
arise from the sheath of a flexor tendon but more frequently derive from the capsule
of the joints between the small carpal bones of the hand. They are more common
in women and may cause dull, aching pain. Among instrumentalists, ganglia of the
flexor sheaths are quite often seen in drummers and tympanists (Winspur and Wynn
Parry 1998). If they do not affect playing they can be left alone; however, if they pre-
vent full extension of the wrist or movement of tendons, or are compressing a nerve,
they will require some form of treatment. This may be as simple as removing the fluid
 The Shoulder, Arm, and Hand 79

they contain with a syringe, though the result is likely to be temporary. Surgery is the
ultimate option, though care must be taken to ensure that neither the tendons nor any
small sensory nerves in their vicinity are damaged. Following surgery, physiotherapy
is required to prevent internal scarring from causing attachments that limit tendon
movement.

Dupuytren’s Contracture
Dupuytren’s contracture is not a disease of tendons, but of a tough fibrous sheet of
tissue (the palmar aponeurosis) that extends across the palm of the hand, from the
flexor retinaculum of the wrist to the base of the fingers. It forms a strong roof above
the flexor tendons, giving them some protection from the compressive forces caused
by gripping. In Dupuytren’s contracture, the fibrous bands of the aponeurosis thicken
and contract, causing the fingers to be drawn into a curled position. It usually starts
with the little and ring fingers and is seen most often in men over fifty, but is almost
exclusively a disease of northern Europeans (Winspur 1998a). It is genetically inher-
ited and may have been spread initially by the Vikings, with the result that it is seen in
many families in Ireland and the west coast of Scotland. It is perhaps most famous as
“the curse of the MacCrimmons,” who were a family of bagpipers from the island of
Skye. The consequences for instrumentalists depend considerably on their instrument.
It causes serious problems for bagpipers because of the importance of the little finger
in playing. It is also troublesome in instruments that require a broad reach, such as
the piano and bassoon, but is much less of a disability in string players, whose fingers
are normally quite flexed during playing. Treatment requires surgery, which must be
carefully managed to ensure a good result but need only entail a break from playing of
one to three weeks with full schedule resuming after one to two months. For a fuller
treatment of the history of this condition, and important guiding principles for surgery,
the reader is referred to Winspur (1998a).

Problems with Joints

Joint Hypermobility
The range of movement that joints are capable of varies between individuals, and
hypermobility is most obviously manifest in what is commonly called double-jointed-
ness. This is frequently encountered among musicians and indeed may even improve
playing facility in instruments where the ability to hold extreme joint angles is an
advantage. Under these circumstances it may even reduce susceptibility to some other
forms of playing-related injury (Rosety-Rodriguez et al. 2003). Historical examples
of individuals with hypermobility include Paganini and Rachmaninoff, but it is actu-
ally more typical of women, especially in the early teenage years, and is particularly
associated with those who have a tall, thin body type (Wynn Parry 1998e). It can
cause considerable problems in joints that support the weight of the instrument or
must remain fixed when strong forces are applied to them (e.g., the wrist in fortissimo
piano playing). In order to reap the benefits without paying the penalty of tendonitis
and sprains, it is important to build up the muscles that support the joints to resist the
80 Chapter Three

forces generated. Splinting may be useful if the muscles are not yet strong enough for
this; for hypermobility of finger joints, ring splints (fig. 3-17C) may be very effective
for guitarists and other string players.

Arthritis
There are two forms of arthritis, osteoarthritis and rheumatoid arthritis. Osteoarthritis
is a disease of aging caused by physical wear of the joint surfaces, and joints that
are used most intensively are likely to be affected first. It frequently manifests itself
in older musicians as pain at the carpometacarpal joint, which lies at the base of the
thumb. This problem is particularly serious for string players in either hand, and of
course, in keyboard players. It also poses problems for the left hand of clarinetists,
though this can be alleviated by using a strap or rod to support the weight of the instru-
ment. There are a number of treatments that can help the musician cope with the con-
dition (Winspur 1998c), for example, gentle stretching of the capsule of an affected
finger or thumb joint or the control of inflammation by injection of anti-inflammatory
drugs into it. If the joint cannot be held in a stable, pain-free position during playing,
small splints may be used for this purpose; as a last resort the joint can be surgically
fixed at an angle that allows playing.
Rheumatoid arthritis is an autoimmune disease in which the body’s defenses against
infection are turned inappropriately against the tissues of the joints. It first causes
swelling in the membranes that secrete the lubricating synovial fluid but may then
progress to a gradual destruction of the cartilages of the joint. The disease often pro-
gresses quite slowly with periodic flare-ups, allowing time to adapt the repertoire to
the limitations it imposes or to alter playing technique to accommodate the resulting
disability. The swelling can be controlled with anti-inflammatory drugs, and surgical
treatment may be beneficial to free tendons that have been immobilized by chronic
inflammation (Wynn Parry 1998d).

Bursitis
If the bursae associated with a joint become inflamed, the surfaces may not slide
smoothly over one another, with the result that movement stretches the tissue, caus-
ing pain. This condition is known as bursitis (the ending “-itis” means inflammation,
so tendonitis is an inflammation of the tendon, arthritis is inflammation of the joint,
appendicitis is an inflammation of the appendix, etc.). In the upper limb, some of the
most commonly affected bursae are those in the shoulder joint (animation 3-2). The
pain may be particularly marked when the arm is raised, a movement required to some
degree by a large proportion of instrumentalists. This movement is brought about by
activity in the deltoid and supraspinatus muscles, both of which lie in contact with
bursae. There are many bursae elsewhere in the body, for example, in the elbow and
knee, and these can also exhibit bursitis. When it develops, movements that aggra-
vate it should be reduced as much as possible and ice packs or drug treatment used
to reduce the inflammation. If the injury is experienced repeatedly, recovery is likely
to become slower, so techniques aimed at its prevention (such as warming up and
stretching exercises) should be cultivated.
 The Shoulder, Arm, and Hand 81

Frozen Shoulder (also Known as Adhesive Capsulitis)

The key to understanding this condition is in the second of its two names. The cap-
sule of the shoulder joint is thin and loose to allow maximum freedom of movement.
The wall of the capsule has folds and creases that can open out when the joint is held
in extreme positions. If the capsule becomes inflamed or torn, scarring may result,
which seals these folds together, restricting the movement of the joint. This may also
happen if the shoulder has been immobilized as a result of another injury, or it may
be associated with problems such as Dupuytren’s contracture. For reasons that are
unknown, it can be a consequence of metabolic disorders such as diabetes or thyroid
conditions. Frozen shoulder may develop during recovery from conditions such as
rotator cuff syndrome when the shoulder is immobilized, for example, with a sling,
for long periods each day. It is therefore important that these periods are interspersed
with gentle stretching exercises that take the arm through its full range of movement
at the shoulder. Treatment for frozen shoulder becomes progressively more difficult
the longer it is delayed. It generally involves physical therapy, ice treatment, and the
use of anti-inflammatory drugs. Though surgery is sometimes advocated, this can
sometimes be counterproductive as it may cause additional adhesions.

Damage to Finger Joints

The proximal interphalangeal (PIP) joint is the middle joint of the finger (fig. 3-7) and
is not only quite prone to damage (generally as a result of nonmusical activities) but
also does not recover well from the inflammation that follows injury. The problem is
that if the joint becomes swollen, it does not remain straight but tends to become fixed
at an angle of 20°–30°. Even when the original injury appears relatively minor (such
as a sprain), this can be difficult to correct because the ligaments around the joint tend
to contract over the first five to ten days after injury. It is therefore important to splint
the joint straight for the first week or so, and subsequently to control the inflammation
and carry out passive and active stretching of the joint in order to regain the full range
of movement (Davies, Rogers, and Kember 1998).

Nerve Problems
Most problems related to the nerves of the arm result from compression, so we will
first consider the structure of the nerve and how this affects its function. Nerves are
composed of bundles of axons (the long, cablelike processes of nerve cells) that
carry information to and from the spinal cord and brain. Some of the axons carry
sensory information from the skin, muscles, and joints, and others run in the oppo-
site direction and control the muscles. The axons are associated with nonneuronal
cells called glial cells, which are vital for their normal function. One type of glial
cell wraps around the larger-diameter axons to form what is called a myelin sheath.
For reasons that we need not go into, this increases the rate at which information
can flow along the axon. If the myelin sheath is damaged, the passage of informa-
tion will slow or stop altogether. Depending on the type of axon affected, this will
lead to either a loss of sensation or to muscle weakness or paralysis. If activity in
82 Chapter Three

the nerve is blocked, the muscles rapidly start to waste away, becoming noticeably
smaller in a matter of weeks.
The axon bundles are surrounded by a fatty, elastic tissue that acts rather like
bubble wrap, protecting them from compression. The entire nerve and its packing are
wrapped in a tough outer sleeve that not only protects the softer tissue underneath
from abrasion, but also adds tensile strength to the nerve. As a limb moves, the large
nerves crossing the joints may slide 1–2 centimeters relative to the surrounding bones
and, at certain points, may be stretched slightly around the bony projections of the
skeleton. This is normal, and the elasticity of the nerve and the padding of fatty tissue
around the nerve bundles protect the axons from damage. Sometimes, however, the
nerve is compressed to a degree that compromises the ability of the axons to func-
tion normally. This may, for example, occur if a joint is held at an extreme angle for
a prolonged period, which stretches it around a bony pulley (e.g., the ulnar nerve at
the elbow). Where nerves run through narrow spaces, such as the carpal tunnel at the
wrist, their ability to slide smoothly during joint movement may be reduced if the
surrounding tissues swell sufficiently to trap the nerve. This can lead them to become
compressed or overstretched or to rub against surrounding structures.
All tissues in the body require a blood supply, and nerves are no exception. One
of the initial effects of nerve compression is that small blood vessels within the nerve
collapse so that the cells within are starved of oxygen, blocking the flow of informa-
tion along the axons. The initial consequence is numbness such as that experienced as
a result of sitting for some time with one leg crossed over the opposite knee, a posture
that compresses a nerve running down the back of the leg. Sensation rapidly returns
when the legs are uncrossed; however, if the compression is maintained for a prolonged
period, the myelin sheath may be damaged. Axons supplying muscles are generally
most vulnerable, so some form of temporary paralysis is usually the main result, and
a return to normal function may take several days. One example of this seen by many
a world-weary general practitioner is “Saturday night palsy,” which results when one
falls asleep in a chair with one arm flung over the back, compressing nerves in the arm-
pit. The name originates from less prosperous times when Saturday was the only night
of alcoholic overindulgence, which not only made sleeping in a chair more likely (no
doubt to keep the marital peace), but also provided the necessary anesthesia to prevent
the victim being awakened by the discomfort. Though actual severing of axons is most
commonly the result of traumatic injury, long-term nerve entrapment can sometimes
also have this result (Lundborg and Dahlin 1994). This leads to long-term loss of sensa-
tion or motor control, which only resolves itself if the axons are able to regrow—a slow
process which at best proceeds at a rate of about one millimeter per day.
One factor which can make precise diagnosis of nerve compression very difficult is
that it is not unusual for it to occur at more than one site on the same nerve (Winspur
1998b; Lundborg and Dahlin 1994). The symptoms may reflect the two primary crush
sites or be typical of only one. Relief of either crush site may (or may not) eliminate
the symptoms of both. The prevalence of double compression suggests that pinching
the nerve in one place makes a second compression more likely. The reasons for this
are unclear, but one possibility is that the first compression induces some swelling
in the nerve as a whole. Another is that trapping the nerve in one place increases the
 The Shoulder, Arm, and Hand 83

likelihood that it will be stretched at another place by joint movement. If entrapment

of a nerve at the wrist indirectly results in compression of the nerve root where it
leaves the vertebral column, the major symptoms may suggest only the problem in
the neck. Surgical relief of this compression (which would be intrusive and risky)
might then reveal the compression at the wrist. If the initial surgery had dealt with the
wrist problem (a much more minor procedure), this might well have resolved the neck
compression. Careful investigation of nerve compression before a definitive diagnosis
is made is therefore very important. The first step involves noting the symptoms of
altered sensation, muscle weakness, and clumsiness. The effects of applying pressure
to the nerves at key points where they may suffer compression should also be deter-
mined. In addition, it is usual to test the speed at which signals run along the nerve
(its conduction velocity) by stimulating it electrically at various points and recording
from both the nerve and the muscles it supplies. This is achieved by placing electrodes
on the skin over the nerve (a nerve conduction study) and recording from muscles us-
ing similar electrodes or by inserting fine wires into them (electromyography). Nerve
compression may then reveal itself as a reduction in the speed of conduction. Though
some patients may feel reluctant to undergo such tests, they use only low voltages and
cause no damage. Though they can be a little uncomfortable, they are very important
in identifying the presence and site of nerve compression and may prevent inappro-
priate surgery. It is important, however, that they be carried out by a clinician who is
experienced in nerve conduction studies, as the results can be misleading if the tests
are not set up properly.
In the arm, it is the median and ulnar nerves that most commonly suffer from
compression, and only these will be considered here. The places where compression
is most likely to occur are shown in figure 3-14. These are the thoracic outlet where
the nerves pass under the collarbone, the ulnar nerve at the elbow, and the ulnar and
median nerves at the wrist.

Thoracic Outlet Syndrome

The nerve roots that emerge from the brachial plexus at the root of the neck to supply
the arm sometimes become compressed. This can either be caused by the muscles they
pass between (the scalenes), by sheets of fibrous tissue that hold muscles and large
blood vessels in place, or occasionally by the presence of an extra rib in the neck,
which is found in about 1 percent of the population. The typical symptoms of thoracic
outlet syndrome are spreading pain down the inner surface of the forearm accompa-
nied by numbness in the little and third finger and a feeling of clumsiness or heaviness
in the limb. These may be particularly associated with certain postures such as the arm
being held out from the body either to the side or in front (Wynn Parry 1998e; Leder-
man 2003), which is a posture typical of many instrumentalists. It should be noted,
however, that some of these symptoms are also typical of compression elsewhere in
the arm. For example, the numbness in the little and third fingers is typical of ulnar
nerve compression, so care must be taken in diagnosis. It is not uncommon to find
that people with certain body types (or somatotypes) are prone to particular physical
problems, and this is true of thoracic outlet syndrome, which is frequently seen in
84 Chapter Three

tall, thin individuals with sloping shoulders, in whom the thoracic outlet is likely to
be narrow. It also appears to be more common in women than in men. The precise
nature and incidence of thoracic outlet syndrome is highly controversial in medicine,
yet it is one of the most common nerve compression problems diagnosed in musicians
(Huang and Zager 2004; Lederman 1987). However, it remains uncertain whether it is
genuinely a condition to which musicians are prone, or whether, because of the rather
general nature of its symptoms, this diagnosis is often made when a more precise
definition of the condition has not been achieved (Winspur 1998b). One of the most
useful indications of genuine thoracic outlet syndrome is said by Wynn Parry to be a
simple provocative test in which the symptoms are reproduced by applying pressure
onto the roots of the brachial plexus at the base of the neck (Wynn Parry 1998b). A
course of shoulder-raising exercises will often resolve the problem, and many musi-
cians find that the Alexander technique is beneficial in improving posture in this
regard. Initial treatment should always be conservative, the object being to strengthen
muscles that resist the tendency for the shoulders to slump downward and compress
the nerve plexus. Physiotherapy leads to a dramatic alleviation of the symptoms in
most cases. The degree to which surgery is used varies considerably between clinics
and is not necessarily linked to the incidence of successful outcome. Due to the ana-
tomical complexity of the region, surgery is quite major and carries significant risks,
though accurate data on how often important nerves or vessels are damaged are hard
to come by (Lederman 1987).

Ulnar Nerve Entrapment

The most common site of entrapment of the ulnar nerve is where it wraps round the
lower end of the humerus (the medial epicondyle). Here it lies in a groove in the bone
called the cubital tunnel, where it is held in place by an overlying fibrous sheet (ani-
mation 3-6). When the elbow is bent the nerve slides within the channel by as much
as a centimeter and is also stretched by several millimeters. The tunnel also becomes
shallower during this movement, compressing the nerve. This is most marked when
the arm is held out from the body and the wrist is extended (Verheyden and Palmer
2002), a posture that is not unusual in musicians (e.g., in violin and viola players). The
compression is most likely to become a problem if the nerve is irritated and inflamed.
The nerve can also be compressed just below in the upper forearm, where it passes
between two parts of one of the flexor muscles of the wrist (flexor carpi ulnaris). In
the early stages of the condition there may be a dull ache at the elbow, followed later
by a numbness and tingling in the little and ring fingers and a loss of grip strength due
to weakness of some of the flexor muscles in the forearm and of the small muscles
within the hand. With the loss of their nerve supply, the muscles in the hand may
start to degenerate. The most obvious sign of this is a flattening of the muscle mass
that moves the little finger and lies along the edge of the hand. The most conservative
treatment is to encourage postures in which the elbow is relatively straight, which,
though impossible for many instrumentalists while playing, can at least be maintained
at other times during the day, and when sleeping. In intractable cases, surgery is often
successful; however, there are several surgical options whose merits need to be care-
 The Shoulder, Arm, and Hand 85

fully compared. Compression of the ulnar nerve is also sometimes seen in the hand,
where it runs through a narrow space between two carpal bones, called Guyon’s canal
(fig. 3-15). This may develop in flutists if the wrists (particularly the left) are habitu-
ally held in a strongly extended position (Wainapel 1988). Initial treatment involves
splinting the wrist to maintain it in a posture that keeps the canal as open as possible;
however, surgical release of the nerve may become necessary.

Median Nerve Entrapment

The most common of all nerve entrapments in the arm is carpal tunnel syndrome,
which is a compression of the median nerve at the wrist (fig. 3-15, animation 3-5);
however, it is also frequently misdiagnosed (Winspur 1998b). Among those who seek
treatment, it is more than twice as common in women as in men (Dawson 1999).
The carpal tunnel is the space under the retinaculum on the palmar side of the wrist
and is crowded with tendons that slide back and forward in their synovial sheaths
as the fingers are moved. Sitting on top of these, just below the retinaculum, is the
median nerve. Inflammation of any of the surrounding structures or tissues can lead
to compression of the nerve. This is sometimes associated with other conditions such
as diabetes or pregnancy, presumably as a result of swelling in the tissue due to fluid
retention. Pressure on the nerve is greatest during extreme flexion or extension of
the wrist, particularly when coupled with finger flexion, a posture common during
playing. In more than half of affected musicians, carpal tunnel syndrome is seen in
both hands, but where this is not the case, it is the right hand that is more commonly
affected, particularly in pianists and orchestral string players, who are the most vulner-
able to carpal tunnel syndrome (Lederman 2003; Dawson 1999). This suggests that
for string players, bowing technique may be a significant causative factor; however,
another possibility is that carpal tunnel syndrome is more common in the dominant
hand, perhaps because of its use in nonmusical activity (Lederman 2002).
Symptoms of carpal tunnel syndrome include tingling or disturbed sensation on the
palmar surfaces of the thumb, index, and middle fingers (but not the palm itself). There
can also be aching at night in the upper arm and shoulder. If the condition is allowed to
progress, there may be a wasting of muscles in the hand supplied by the median nerve.
The most dramatic indication of this is a thinning of the large pad of muscles called
the thenar eminence, which controls the thumb. If the condition is of recent origin and
not severe, treatment involves splinting of the wrist combined with anti-inflammatory
drugs; however, in a considerable proportion of cases, surgery is necessary, though
great care must be taken not to damage the nerve during this procedure.

Entrapment of Small Nerves in the Fingers

The main nerve branches that supply the fingers and the thumb run up their sides,
where they may be vulnerable to compression in some instruments. For example,
in the left hand of flutists the instrument rests partly on the side of the first segment
(phalanx) of the index finger, just below the PIP joint, where it may rest on the digital
nerve. Digital nerves may also be damaged when the side of the finger is used to pluck
the string (e.g., in harpists or guitarists).
86 Chapter Three

PREVENTING INJURY—
THE APPROACH TO PLAYING AND PRACTICE

Practice
For most musicians, a significant proportion of playing time is spent in the daily
routine of individual practice, which is necessary to maintain concert levels of per-
formance and to learn new repertoire. These periods are under the complete control
of the player and can be carefully planned to develop and maintain a technique that
is not only effective at producing the required musical results, but will also limit the
chance of injury. There should be no conflict of interest between these aims because
the two are usually synonymous. Useful information on many aspects of practice and
posture and minimizing tension is available from several sources (Wynn Parry 1998a;
Warrington 2003; Horvath 2002; Culf 1998; Paull and Harrison 1997). The first stage
is to warm up the muscles to be used, something which is done without the instrument.
This is particularly important in the morning when muscles are stiffer and tighter af-
ter the hours of inactivity spent in sleep. Warm-up is best followed by the playing of
some easy pieces or exercises concentrating on accuracy, at the same time settling into
a comfortable and sustainable posture (about which, more below). The room should
be at a reasonable temperature; if this is not possible, the arms and hands need to be
kept warm by wearing appropriate clothing in order to maintain the temperature and
suppleness of the muscles. This is particularly important for the forearms, which con-
tain the large muscles moving the hand and fingers. Some players may find it useful
to wear fingerless gloves though even with these, cold fingers may tend to slip on the
keys because of the lack of moisture on the pads from sweat.
Prolonged periods of practice should be split up into short sessions of perhaps
thirty-five to forty-five minutes separated by intervals of a few minutes during which
it is beneficial to do some stretching exercises for the playing (arm and hand) and
postural (back, neck, and shoulder) muscles. After finishing playing, a further period
of stretching is useful as a way of preventing the asymmetries of playing posture
being carried over into normal life. During this period you will probably want to
concentrate on the postural muscles that have been active but produce relatively
little movement—those in the back, neck, shoulders, and upper arm. The demands
of playing are similar to those of more traditionally recognized athletic pursuits, the
only difference being that many of the muscles most actively employed are rela-
tively small. In order to prevent strain, they should be given the same attention as
those used in sports. The athletic analogy applies equally to how you approach an
increase in practice intensity. An athlete switching from speed to strength training or
vice versa will make the transition gradually. Similarly, if as a musician you intend
to increase the amount of time you play each day, this must be built up gradually.
If you have been on holiday for a couple of weeks during which you did little or no
playing, it is not sensible to return immediately to the same schedule you were fol-
lowing immediately before the break. This is even more relevant if you are returning
to orchestral or group playing after injury. We can all see that this would be common
sense for a soccer player or runner, but it somehow it appears less obvious for musi-
cians, though it is just as relevant.
 The Shoulder, Arm, and Hand 87

The demands made of the body may also change if you switch to a new instrument.
Changing from a flute with a C foot joint to one with a B foot joint will significantly
increase the weight you must support, as will exchanging a single French horn for a
double. Among string players, quite small increases in bow weight can create signifi-
cant arm problems if the change is not managed carefully, while instruments such as
violas can vary considerably in their dimensions. Perhaps after learning your craft as
a woodwind player on a modern instrument, you decide to explore the baroque rep-
ertoire with a period instrument. Though the instrument itself will be lighter, it may
require a greater finger spread. In all of these situations, it is sensible to transfer to the
new instrument gradually to give yourself time to adapt physically to the change of
loading on the limbs. It is usually advisable to return to quite simple repertoire during
the initial period of adjustment.
Many players attribute great importance to mental rehearsal (see chapter 8), which
means going through a piece from the score or from memory while away from the
instrument, rehearsing mentally the movements of the fingers and arms (Fischer 1998;
Warrington 2003). Mental rehearsal is also common among athletes who perform
complex body movements, such as divers, gymnasts, or high jumpers, and more will
be said about this topic in the chapter dealing with the control of motor activity by the
brain. For musicians the advantage of mental rehearsal is that it allows familiarization
with the structure of the piece, virtual practice of movements, and even the develop-
ment of interpretation when access to an instrument is not possible and without the
risk of physical strain. It may also help to reveal sources of muscle tension, which may
be perceptible during mental rehearsal of some note sequences even in the absence of
full playing movements of the fingers.
In an orchestral or ensemble environment, the player has less control over the struc-
ture of the playing sessions than when practicing alone; however, even in the group
setting, discreet stretching exercises and judicious strategies for resting overtaxed
joints can be practiced in the intervals between bouts of playing. An extensive range
of rest and stretching techniques developed for use on stage and off has been compiled
by Janet Horvath, a professional cellist who has suffered from and overcome severe
overuse problems (Horvath 2002).

Posture and Movement

The details of correct posture for the whole gamut of orchestral and nonorchestral
instruments are beyond the scope of this book; however, we will here deal with the
general anatomical principles that underlie good arm posture. Back posture is dealt
with in chapter 2, and information relevant to individual instruments is found else-
where (Horvath 2002; Wynn Parry 1998a; Paull and Harrison 1997).
Over the last two decades, there have been a considerable number of studies di-
rected at analyzing the postures and movements that underlie the playing of keyboard
and string instruments. For bowed stringed instruments in particular it has proved
much easier to collect data than to find a useful way of analyzing them and applying
the results, which may explain the high proportion of “preliminary studies” in the liter-
ature. Though equipment is now available to quantify the complex three-dimensional
88 Chapter Three

movements required in string playing, analyzing sufficient numbers of individuals

with different levels of experience and different body types to determine what might
be optimal strategies remains highly challenging. Almost all of the published literature
sets out with the stated aim of providing information that will help players to avoid
injury (Visentin and Shan 2003, 2004; Shan and Visentin 2003; Turner-Stokes and
Reid 1999), but this remains a distant goal. The range of variables relevant to this type
of analysis is daunting and many, such as the dimensions of the players themselves,
have yet to be considered. The best way forward may be to focus on relatively nar-
row but well-defined questions of likely biomechanical or clinical significance, such
as the angle or loading of a particular joint. One of the earliest investigations into the
physiology of string playing published in English is that of Szende and Nemessuri
(Szende and Nemessuri 1971), which concentrates mainly on providing an accurate
description of which muscles are active during various playing movements. Given the
technology of the day this was a considerable achievement, and in addition it showed
a general awareness of the roots of performance-related injury well ahead of its time.
There have been few attempts to record muscle activity during playing since, though
two brief papers by Koehler (1995, 1996) suggest that this type of information might
be useful in biofeedback strategies for optimizing muscle activity.
One of the joints in the arm most critical in terms of posture is the wrist. It is vul-
nerable because so many tendons and nerves cross it in the narrow channels that lie
beneath the restraining bands of the retinacula. This space available is greatest (and
the potential friction on tendons least) when the wrist is held in a neutral position as
a straight extension of the arm. In this position it is neither extended or flexed, nor
deviated to the radial (thumb) or ulnar (little finger) side. When playing an instrument
requiring rapid finger movement, the wrist should be kept close to this position as
often as is possible. One group of instrumentalists who are particularly prone to hold
both the left and the right wrists in a hyperflexed position are guitarists (fig. 3-18).
One reason for this may be that they are often self-taught, but it is not difficult to avoid
this playing posture. Players of bowed string instruments also need to guard against
maintaining a strong flexion of the wrists in either hand, though it is unavoidable at
the farthest reaches of the fingerboard, where strong ulnar deviation is also necessary.
In pianists both extreme flexion and extension of the wrist are sometimes seen as a
result of using a piano stool that is either too low or too high (animation 2-3). If set
at a height that allows the forearms to lie close to the horizontal, the wrist will return
to the neutral position. The shoulders should be kept relaxed as much as possible,
though tension may be used periodically to create certain effects during playing. The
pianist should also guard against excessive radial and ulnar deviation of the wrist (fig.
3-19). Avoiding this may periodically require the player to extend the arm away from
the body, to lean to one or other side, or even to move along the stool for complex
passages at the extreme ends of the keyboard. For most wind instruments the wrists
naturally lie in the correct position; however, in the flute there is often a tendency to
hyperextend the left wrist and to flex the right wrist. Extension of the left wrist can
be reduced by clipping a support such as a Bo-Pep onto the instrument. If you need
to convince yourself of the advantages of the neutral wrist position, try this experi-
ment, suggested by Janet Horvath (2002). Spread and straighten the fingers, first with
 The Shoulder, Arm, and Hand 89

Figure 3-18. Examples of good and bad wrist and hand position in a guitarist. Good technique
involves reducing wrist flexion to a minimum. In the poor posture, both wrists are strongly flexed,
which may compress the tendons and median nerve.

the wrist fully extended (cocked up), then with the wrist fully flexed (drooping), and
finally in the correct horizontal position. The difference in the degree of comfort and
in the force required for the movement is quite remarkable.
When developing good technique, the ultimate aim should be to eliminate any un-
necessary muscular activity from playing. This has been investigated in a study that
compared the forces applied to the keys by amateur pianists (who played less than
one hour a day) and expert players (who practiced four hours or more a day) (Parlitz,
Peschel, and Altenmuller 1998). Interestingly, in the photograph showing the equip-
ment used to monitor key force, the hand of the pianist exhibits extreme wrist exten-
sion. One hopes that this belonged to a member of the amateur group! Several of the
90 Chapter Three

Figure 3-19. Ulnar deviation of the wrist in pianists. A. When playing in the center of the key-
board, the forearms point inward and the fingers do not lie parallel to the keys. B. To bring the
fingers in line with the keys, the wrist must be rotated outward (ulnar deviation). This may restrict
the movement of finger tendons in the carpal tunnel. C. If the elbows are held away from the
body, the misalignment of fingers leads to an even more pronounced ulnar deviation.

tasks (taken from Dohnanyi studies) on which the comparison was based required
one or more keys to be held down while the remaining fingers played a repeating
pattern of eighth notes as in a very slow trill. The expert players applied a very light
and unvarying pressure to the keys of the sustained notes, while the pressure applied
by the amateurs was many times greater and fluctuated with the movement of the
moving fingers. The amateurs were largely unaware of the force they were exerting,
which appeared to form part of a strategy aimed at stabilizing the stationary fingers.
This was not necessary for the experienced pianists because they had developed a
much more independent control of the fingers; however, even among such players
a continuing reduction in finger force with the number of years of playing has been
found (Wolf et al. 1993; Jabusch 2006). In more difficult tasks of a similar nature
set by Parlitz et al., the expert players did increase the force on the stationary fingers
slightly. Although this time there was less difference between the two groups in the
magnitude of the pressure applied by the moving fingers, the experts applied this for
much shorter periods during each key strike. When playing passages with a regular
rhythm, for example, sixteenth-note scales, the variability in the time between each
note strike declines markedly with experience, though this may require many years to
develop (Wagner, Poingek, and Teckhaus 1973).
The forces acting on the individual joints of the fingers are affected by the posture of
the hand. They are at a maximum when the hand is held flat with the fingers extended
straight out. When the fingers are curved so that the last joint lies perpendicular to the
key, the forces on the last finger joint (distal interphalangeal joint or DIP; fig. 3-7)
and on the tendon of the flexor digitorum profundus muscle (which moves that joint)
are at a minimum (Wolf et al. 1993). A little self-experimentation will reveal that one
negative consequence of curving the fingers is to make abduction and adduction at
the metacarpophalangeal (knuckle) joint (i.e., spreading the fingers and drawing them
together again) more difficult (Bejjani et al. 1989), as is circumduction (moving the
finger in an arc). In this, the index finger is least affected. Hand positions that theoreti-
cally minimize the force applied to all of the finger joints involve moderate flexion
of the metacarpophalangeal joint (Harding, Brandt, and Hillberry 1993), which would
 The Shoulder, Arm, and Hand 91

necessitate considerable extension of the wrist. Optimizing hand position will always
therefore remain a compromise (Harding, Brandt, and Hillberry 1998).
Joints are usually moved by pairs of antagonistic muscles; for example, there are
muscles that flex the fingers, and this movement can be opposed by muscles that ex-
tend them. The differences seen between the two groups of pianists mentioned above
may partly have been the presence in the amateurs of a significant degree of simul-
taneous activity (co-contraction) in antagonistic muscle pairs controlling the fingers.
One might imagine that when the flexors are contracting, the extensors would be re-
laxed and vice versa; however, this is often not the case. To take an extreme example,
if a weight lifter holds a barbell above his or her head, the flexors and extensors of
the wrist contract simultaneously to hold it straight, thus preventing the weight from
turning the joint in either direction. Where movement is required, co-contraction can
still be helpful in promoting joint stability. For the fingers of a pianist, for example,
this would apply when the keys are struck forcibly in very loud passages. However,
in most playing situations, it is neither necessary nor helpful and simply slows finger
speed and responsiveness. Increasing the pressure on a string or on the key of a wood-
wind or brass instrument above the level needed for a firm contact is pointless. If you
are unsure of the notes you are playing, however, you may find you are co-contracting
opposing finger muscles, perhaps in unconscious readiness to move the finger in the
opposite direction if a mistake is made. In order to relax them, the notes may have to
be relearned securely. Without constant vigilance, this habit can become ingrained.
Whatever its reason, co-contraction makes moving the joint in the correct direction
harder, as it is necessary to overcome the opposing muscle to do so. It may also cause
the fingers to rise higher than necessary when taken off the key, as a greater initial
force is needed to overcome the continuing flexor muscle activity. This extends the
minimum period before the finger can be brought back into contact with the key, slow-
ing playing speed. As you become more familiar with a piece of music, co-contraction
is likely to decline (Fry 2000); nevertheless, a conscious effort to relax antagonistic
muscles should be maintained at all times so that only the minimum force is used.
Not attempting to play complex passages at full speed until the notes are fully secure
may also help eliminate co-contraction and in the long run shorten the time required
to play these passages fluently.
Stress or anxiety also induces inappropriate muscle tension. If you are grinding
your teeth at night when you are asleep, you are probably co-contracting your finger
muscles by day as well as experiencing tension in your shoulders and arms. Under
such circumstances, playing becomes very tiring! This is one reason why stress, which
is basically a psychological factor, can produce the physical symptoms of overuse
injury.

Instrument Ergonomics
The shape and form (the somatotype) of the human body varies greatly between indi-
viduals; however, for reasons dictated largely by acoustics and tradition, most instru-
ments come in one size only. Musicians generally choose to play a particular instrument
for emotive reasons related to the quality of the sound it produces, and only rarely alter
92 Chapter Three

their choice because it does not suit their physical attributes. Of course, through thou-
sands of hours of playing the body will probably become modified to a considerable
degree to accommodate the requirements of the instrument. Development of increased
breath control, physical strength and stamina, and neuromuscular coordination are all
part of learning to play an instrument well. Ultimately, however, hands or arms do not
grow larger or bodies taller because a pianist or cellist requires it. By developing great
technique, some players can compensate wonderfully for what appear to be major physi-
cal disadvantages; however, there are obvious limits to this process. In attempting to
achieve mastery over the instrument, temporary gain may be acquired only as the result
of adopting postures or habits that are in the long term unsustainable. The physical chal-
lenges posed by the instrument are not restricted only to playing. Many instruments are
bulky or heavy (particularly in their protective cases), and some of the most awkward,
such as the harp, are traditionally played by the smaller members of the orchestra. The
risk of injury when moving an instrument may be alleviated by something as simple as
adding wheels to the case. While this may be standard for very large or awkward instru-
ments, smaller or younger players should consider installing these for instruments such
as the cello or horn, for which they may currently be less common.
It is a strange anomaly that many musical instruments do not appear to have been
designed with the human body in mind but more in accordance with certain aesthetic
principles. Sometimes even the posture used in playing is influenced by art! Prinner
in 1677 advised players to hold the violin firmly with the chin in order to increase the
dexterity of the left hand, which as a result could be released from having also to sup-
port the instrument. Apparently some players of the time rested the instrument against
the chest or arm, not because they felt that this was more comfortable or produced
better results, but in imitation of a painting of an angel playing to St. Francis (Manze
2003). The picture in question is probably one by Guercino (1591–1666), now housed
in the Gemäldegalerie of Dresden, showing an angel in just such a posture, which is
more akin to that used for the viola da braccia.
Other problems arise when instruments evolve from a form that was once quite
manageable to one that now puts considerable strain on the body. The early clarinet,
which was made of boxwood and had a relatively simple set of keys (Baines 1977),
weighed only about 10.5 ounces (300 grams). A good modern instrument made of
Grenadilla wood, with the Boehm key system, or indeed a plastic student instrument,
weighs 1.75–2 pounds (800–900 grams), but the player is still expected to support the
weight on the end of the right thumb. This is frequently a source of discomfort. No
one would dream of attempting to play one of the larger woodwind instruments such
as an alto saxophone or bassoon while supporting the instrument solely in this way,
but until recently there appears to have been little impetus to develop alternatives
for smaller instruments such as the clarinet or oboe. Fortunately this is beginning to
change (Markison 1990, 1998; Wynn Parry et al. 1998), and a variety of supportive
devices have been invented by players and instrument makers. These include neck
slings, supporting rods, or modifications of the existing support to spread the weight
over a larger area or to transfer the weight to the metacarpophalangeal joint at the
base of the thumb. Such supports may be of particular interest to older players, as the
joints of the thumb are frequently among the first to be affected by arthritis. Even in
 The Shoulder, Arm, and Hand 93

the absence of physical problems, many players find it useful to employ such supports
and indeed may use a range of them in various circumstances. A full discussion of the
ever-increasing range of such supports is beyond the scope of this book, but a review
of some of these can be found in Rosset i Llobet and Odam (2007). I will illustrate the
principles behind ergonomic woodwind design by considering a progressive series of
adaptations that are used in the flute, which, among woodwind instruments, is one of
the most problematical for posture. Much of the difficulty stems from the necessity of
holding the instrument horizontally out to the right. This requires that the right arm be
held abducted and the left hand, strongly supinated. In order to align the mouth with
the lip plate, the head and trunk must be rotated to the left, with the result that the left
arm is held across the body. In the alto flute, the head joint doubles back on itself in
a “U” shape so that the right arm can comfortably reach the keys, and similar head
joints for the standard flutes may be useful for children. Those interested in what ef-
fect this has on the acoustics of the instrument are referred to Coltman (1984). Other
options include changing the angle of the head joint relative to the body of the instru-
ment so that the right arm lies lower, or in the most extreme case, so that the body
of the instrument runs vertically like that of other woodwind instruments, though this
may create other ergonomic problems. A number of less extreme modifications can be
made to the instrument to ease the position of the left wrist and hand. Though in many
instruments the keys played by the first three fingers of the left hand lie in line (fig.
3-20A), most manufacturers now offer an offset G option (fig. 3-20B) that reduces the
degree of supination required to place the ring finger on its key. Some players find it
useful to extend the length of all of the keys played by the first three fingers, though
this is not a standard configuration (fig. 3-20D). An inexpensive plastic saddle can
also be clipped to the instrument to reduce the pressure needed to support it by the left
hand, reduce the extension of the left wrist, and allow the index finger to adopt a less
cramped position (fig. 3-20C).
The major postural problems posed by brass instruments stem largely from their
weight. Over the last few decades this has increased considerably for many instru-
ments due to changes in design in response to the demand for a greater volume of
sound. In the last chapter we saw how this can cause problems for the muscles of the
shoulder and the roots of the brachial plexus as they emerge from between the scalene
muscles. A set of support devices (Ergobrass) are available that are based on tele-
scopic rod supports fixed to the instruments. These run either to the floor, to the seat of
the chair between the player’s legs, or to a sling that can be worn discreetly under the
jacket. These supports may also be useful for young players, as, by freeing them from
the need to support the instrument, they will be able to adopt a good playing posture.
They can be gradually weaned off the support as they develop physically.
String players may be more resistant to tampering with the structure of their instru-
ments than wind players; however, they may want to bear in mind that many of the
priceless historical instruments now in circulation are far from their original state.
That Stradivarius cello you covet did not acquire its spike until the nineteenth century,
but the thought of gripping the cello between the shins in the pursuit of authenticity is
not an idea that many modern players would relish and would impose a severe limit
on the use of vibrato. Cellists and double-bass players also have an option for bows
Figure 3-20. A progressive series of ergonomic modifications of the flute. A. In-line G keys (ar-
rows) require maximal supination of the left wrist so that the upper of the two can be covered by
the pad of the ring finger. B. By offsetting the G keys, the requirement for supination is reduced.
C. Addition of a support for the basal segment of the index finger (e.g., a Bo-Pep) reduces wrist
extension and index finger flexion. D. Extending the keys operated by the little, ring, and middle
fingers can further reduce wrist supination.
 The Shoulder, Arm, and Hand 95

between the more commonly used French type (which is similar to the modern violin
bow) and the German type. The German bow has several advantages for those expe-
riencing arm problems. It requires less inward rotation of the shoulder and flexion or
pronation of the wrist and reduces the force transmitted through the basal joint of the
thumb (fig. 3-21A, B). Though ergonomic factors are not the sole concern for players,
this may provide a solution if problems with these joints prove otherwise intractable.
If you are lucky enough to be the owner of a seventeenth- or eighteenth-century
violin or viola and are playing in a conventional orchestra, your instrument will have
undergone an even more drastic transformation to increase the volume of sound it
produces. The original neck will have been replaced by one that is longer and thinner
and has been reset at a greater angle relative to the body. If this is the case, the original
scroll may have been grafted back on. It will have acquired a new, longer fingerboard,
the bridge will be higher, and even the bass bar that lies on the inside of the belly will
have been replaced by one that is thicker and longer than the original. The strings,
some of which are now overwound, are at higher tension, and the instrument is played
with a different type of bow. In addition, a chin rest and shoulder rest will have been
added. From this perspective, a few further changes from the traditional design may
not seem so sacrilegious if you are in the market for a modern instrument. Remember,
if the instrument is more user friendly, not only will you reduce your susceptibility
to injury, you are also likely to play better. The physical challenges for the player are
greater in the viola than the violin due to its larger (and more variable) size. Recently
a number of approaches have been proposed to improve its ergonomics. Otto Erdesz
created some instruments in which the upper bout on the treble side was cut away to
improve access to the fingerboard (fig. 3-21E), and recently more radical solutions
have emerged from the designs of David Rivinus. These have addressed not only the
ergonomics but also the acoustics of the instrument. The least extreme of these dis-
places and tilts the fingerboard toward the treble side to improve the access of the left
hand (fig. 3-21D). In Rivinus’s Pellegrina viola, however, the symmetry of the body
of the instrument has been sacrificed to improve ergonomics and sound quality, and
though the result may seem to owe more than a little to Salvador Dali, it is clearly an
instrument designed with the body of the player in mind (fig. 3-21F). The fingerboard
is again displaced and canted, and the rib height on the bass side is higher than that
on the treble. In addition, the body of the instrument is distorted asymmetrically so
that the area of the belly is increased without affecting access to the fingerboard or
increasing the distance from the button to the peg box. The result of these and other
modifications is to produce an instrument that sounds like a large viola, but which is
ergonomically more friendly than a small one.
Notwithstanding some flirtations with 7/8-size keyboards (Wristen et al. 2006), the
options for pianists are limited, as the dimensions of piano keys are necessarily stan-
dardized, though players of baroque music benefit not only from the narrower keys
of the harpsichord, but also from music that demands less in terms of hand extension
and in finger force, which in these instruments plays no role in dynamics. However,
the weight of the action will vary from instrument to instrument, and though for re-
cital purposes one must accept the piano provided, the player’s own piano used for
the long hours of daily practice at home is another matter. This can be regulated and
Figure 3.21. The two types of cello bows and the different hand grip used for each. A.
French type. B. German type. Note that the hand is pronated and the wrist flexed for
the French bow, but supinated and straighter for the German. C–F. Ergonomic modi-
fications to the viola. C. A normal viola. D. Design by David Rivinus; the shape of the
body is unchanged but the neck is moved to one side to improve the access of the left
hand to the upper parts of the fingerboard. E. Design by Otto Erdesz; the upper bout
is cut away to improve left-hand access. F. The Pellegrina viola is a radical ergonomic
redesign by David Rivinus (www.rivinus-instruments.com/) in which the symmetry of
the body of the instrument is sacrificed for practical aims. The ribs on the bass side
(inset) are also higher than those on the treble side.
 The Shoulder, Arm, and Hand 97

maintained to ensure that the action is both light and even across the keyboard, which
will help to reduce the chances of injury.

REFERENCES

Allieu, Y., K. Hamitouche, J. L. Roux, and Y. Beaton. 1998. Unique surgical conditions. In The
musician’s hand: A clinical guide, ed. I. Winspur and C. B. Wynn Parry. London: Martin
Dunitz.
Baader, A. P., O. Kazennikov, and M. Wiesendanger. 2005. Coordination of bowing and fin-
gering in violin playing. Brain Res Cogn Brain Res 23 (2–3):436–43.
Baines, A. 1977. Woodwind instruments and their history. London: Faber and Faber.
Baker, D. S., J. S. Gaul Jr., V. K. Williams, and M. Graves. 1981. The little finger
superficialis—clinical investigation of its anatomic and functional shortcomings. J Hand
Surg (Am) 6 (4):374–78.
Beauchamp, R. 2003a. Curved fingers—and tension. Music and Health, at www.musicand
health.co.uk/articles/tension.html (accessed April 23, 2008).
———. 2003b. Examples of passages that may cause problems due to tendon linkages or ab-
sences. Music and Health, at www.musicandhealth.co.uk/linkages.htm (accessed April 23,
2008).
Bejjani, F. J., L. Ferraram, N. Xu, C. M. Tomaino, L. Pavlidis, J. Wu, and J. Dommerholt.
1989. Comparison of three piano techniques as an implementation of a proposed experimen-
tal design. Med Prob Perform Artists 4:109–13.
Bergman, R. A., A. K. Afifi, and R. Miyauchi. 2004. Illustrated encyclopedia of human
anatomic variation, at www.anatomyatlases.org/AnatomicVariants/AnatomyHP.shtml (ac-
cessed April 23, 2008).
Brandfonbrener, A. G. 2003. Musculoskeletal problems of instrumental musicians. Hand Clin
19 (2):v–vi, 231–39.
Brown, S. 2000. Promoting a healthy keyboard technique. In Medical problems of the instru-
mental musician, ed. R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Coltman, J. W. 1984. Enhanced power from a recurved flute. J Acoust Soc Am 75:1642–43.
Culf, N. 1998. Musician’s injuries: A guide to their understanding and prevention. Guildford,
UK: Parapress.
Davies, A. T., G. Rogers, and J. Kember. 1998. The physical therapist’s contribution. In The
musician’s hand: A clinical guide, ed. I. Winspur and C. B. Wynn Parry. London: Martin
Dunitz.
Dawson, W. J. 1999. Carpal tunnel syndrome in instrumentalists: A review of 15 years’ clinical
experience. Med Prob Perform Artists 14 (1):25–29.
Fischer, S. 1998. Technique and ease in violin playing. In The musician’s hand: A clinical
guide, ed. I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
Fishbein, M., S. E. Middlestadt, V. Ottati, S. Straus, and A. Ellis. 1998. Medical problems
among ICSOM musicians: Overview of a national survey. Med Prob Perform Artists 3
(1):1–8.
Franco, G. 1999. Ramazzini and workers’ health. Lancet 354 (9181):858–61.
Fry, H. J. H. 2000. Overuse syndrome. In Medical problems of the instrumental musician, ed.
R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Fry, H. J., and G. Rowley. 1992. Instrumental musicians showing technique impairment with
painful overuse. Md Med J 41 (10):899–903.
98 Chapter Three

Harding, D. C., K. D. Brandt, and B. M. Hillberry. 1993. Finger joint force minimization in
pianists using optimization techniques. J Biomech 26 (12):1403–12.
———. 1998. Minimization of finger joint forces and tendon tensions in pianists. Med Prob
Perform Artists 4:103–8.
Hochberg, F. H., R. D. Leffert, M. D. Heller, and L. Merriman. 1983. Hand difficulties among
musicians. JAMA 249 (14):1869–72.
Horvath, J. 2002. Playing (less) hurt: An injury prevention guide for musicians. Minneapolis,
Minn.: J. Horvath.
———. 2004. Stressing prevention. Symphony (March–April).
Huang, J. H., and E. L. Zager. 2004. Thoracic outlet syndrome. Neurosurgery 55 (4):897–902;
discussion 902–3.
Jabusch, H. C. 2006. Movement analysis in pianists. In Music, motor control and the brain, ed.
E. Altenmuller, M. Wiesendanger, and J. Kesselring. Oxford: Oxford University Press.
Kloeppel, R. 2000. Do the “spreadability” and finger length of cellists and guitarists change
due to practice? Med Prob Perform Artists 15:23–30.
Koehler, W. K. 1995. The effect of electromyographic feedback on achievement in bowing
technique. Proceedings of the Second International Technological Directions in Music
Learning Conference, at music.utsa.edu:16080/tdml/conf-II/II-Koehler.html (accessed April
23, 2008).
———. 1996. Physiological mapping of bowing: Six case studies. Proceedings of the Third
International Technological Directions in Music Learning Conference, at music.utsa.edu/
tdml/conf-III/III-Koehler/III-Koehler.html (accessed April 23, 2008).
Lederman, R. J. 1987. Thoracic outlet syndromes: Review of the controversies and a report of
17 instrumental musicians. Med Prob Perform Artists 2:87–91.
———. 2002. Neuromuscular problems in musicians. Neurologist 8 (3):163–74.
———. 2003. Neuromuscular and musculoskeletal problems in instrumental musicians. Muscle
Nerve 27 (5):549–61.
Lederman, R. J., and C. H. Calabrese. 1986. Overuse syndromes in instrumentalists. Med Prob
Perform Artists 1:7–11.
Leijnse, J. N., J. E. Bonte, J. M. Landsmeer, J. J. Kalker, J. C. Van der Meulen, and C. J. Sni-
jders. 1992. Biomechanics of the finger with anatomical restrictions—the significance for
the exercising hand of the musician. J Biomech 25 (11):1253–64.
Leijnse, J. N., C. J. Snijders, J. E. Bonte, J. M. Landsmeer, J. J. Kalker, J. C. Van der Meulen,
G. J. Sonneveld, and S. E. Hovius. 1993. The hand of the musician: The kinematics of the
bidigital finger system with anatomical restrictions. J Biomech 26 (10):1169–79.
Leijnse, J. N., E. T. Walbeehm, G. J. Sonneveld, S. E. Hovius, and J. M. Kauer. 1997. Connec-
tions between the tendons of the musculus flexor digitorum profundus involving the synovial
sheaths in the carpal tunnel. Acta Anat (Basel) 160 (2):112–22.
Lippman, H. I. 1991. A fresh look at overuse syndrome in musical performers: Is “overuse”
overused? Med Prob Perform Artists 6:57–60.
Lundborg, G., and L. B. Dahlin. 1994. Pathophysiology of nerve compression. In Repetitive
motion disorders of the upper extremity, ed. S. L. Gordon, S. J. Blair, L. J. Fine, and K. N.
An. Rosemont, Ill.: American Academy of Orthopaedic Surgeons.
Manze, A. 2003. Biber violin sonatas (sleeve notes). Harmonia Mundi, HMX 2907344.45.
Markison, R. E. 1990. Treatment of musical hands: Redesign of the interface. Hand Clin 6
(3):525–44.
———. 1998. Adjustment of the musical interface. In The musician’s hand: A clinical guide,
ed. I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
 The Shoulder, Arm, and Hand 99

Miller, G., F. Peck, A. Brain, and S. Watson. 2003. Musculotendinous anomalies in musician
and nonmusician hands. Plast Reconstr Surg 112 (7):1815–22; discussion 1823–24.
Miller, G., F. Peck, and J. S. Watson. 2002. Pain disorders and variations in upper limb mor-
phology in music students. Med Prob Perform Artists 17:169–72.
Newmark, J., and F. H. Hochberg. 1987. “Doctor, it hurts when I play”: Painful disorders
among instrumental musicians. Med Prob Perform Artists 2:93–97.
Newmark, J., and R. J. Lederman. 1987. Practice doesn’t necessarily make perfect. Incidence
of overuse syndromes in amateur instrumentalists. Med Prob Perform Artists 2:93–97.
Parlitz, D., T. Peschel, and E. Altenmuller. 1998. Assessment of dynamic finger forces in pia-
nists: Effects of training and expertise. J Biomech 31 (11):1063–67.
Paull, Barbara, and Christine Harrison. 1997. The athletic musician: A guide to playing without
pain. Lanham, Md.: Scarecrow Press.
Perkins, R. E., and M. H. Hast. 1993. Common variations in muscles and tendons of the human
hand. Clin Anat 6:226–31.
Ramazzini, B. 1964. Diseases of workers, History of medicine series, 23. New York: Hafner
Publishing.
Reilly, K. T., M. A. Nordstrom, and M. H. Schieber. 2004. Short-term synchronization between
motor units in different functional subdivisions of the human flexor digitorum profundus
muscle. J Neurophysiol 92 (2):734–42.
Rosety-Rodriguez, M., F. J. Ordonez, J. Farias, M. Rosety, C. Carrasco, A. Ribelles, J. M.
Rosety, and M. Gomez del Valle. 2003. The influence of the active range of movement of
pianists’ wrists on repetitive strain injury. Eur J Anat 7 (2):75–77.
Rosset i Llobet, J., and G. Odam. 2007. The musician’s body: A maintenance manual for peak
performance. London: Guildhall School of Music and Drama; Burlington, Vt.: Ashgate
Publishing.
Rozmaryn, L. M. 1993. Upper extremity disorders in performing artists. Md Med J 42
(3):255–60.
Sataloff, R. T., A. G. Brandfonbrener, and R. J. Lederman. 1998. Performing arts medicine.
2nd ed. San Diego: Singular Publishing Group.
Schmied, A., C. Ivarsson, and E. E. Fetz. 1993. Short-term synchronization of motor units in
human extensor digitorum communis muscle: Relation to contractile properties and volun-
tary control. Exp Brain Res 97 (1):159–72.
Schmied, A., S. Pagni, H. Sturm, and J. P. Vedel. 2000. Selective enhancement of motoneurone
short-term synchrony during an attention-demanding task. Exp Brain Res 133 (3):377–90.
Shan, G., and P. Visentin. 2003. A quantitative three-dimensional analysis of arm kinematics
in violin performance. Med Prob Perform Artists 18:3–10.
Silver, J. K., and L. M. Rozmaryn. 1998. Overuse tendinitis of the intrinsic muscles. Orthope-
dics 21 (8):891–94.
Stotko, L. 1998. Dynamically splinting the musical hand. Med Prob Perform Artists 13:109–13.
Szende, O., and M. Nemessuri. 1971. The physiology of violin playing. London: Collet’s.
Turner-Stokes, L., and K. Reid. 1999. Three-dimensional motion analysis of upper limb
movement in the bowing arm of string-playing musicians. Clin Biomech (Bristol, Avon) 14
(6):426–33.
Verheyden, J. R., and A. K. Palmer. 2002. Cubital tunnel syndrome. E-medicine, at www
.emedicine.com/orthoped/topic479.htm (accessed April 23, 2008).
Visentin, P., and G. Shan. 2003. The kinetic characteristics of the bow arm during violin
performance: An examination of internal loads as a function of tempo. Med Prob Perform
Artists 18:91–97.
100 Chapter Three

———. 2004. An innovative approach to understand overuse injuries: Biomechanical model-

ing as a platform to integrate information obtained from various analytical tools. Med Prob
Perform Artists 19:90–96.
Wagner, C., E. Poingek, and L. Teckhaus. 1973. Piano learning and programmed instruction.
J Res Music Educ 21 (2):106–22.
Wainapel, S. F. 1988. The not-so-magic flute: Two cases of distal ulnar nerve entrapment. Med
Prob Perform Artists 3:63–65.
Warrington, J. 2003. Hand therapy for the musician: Instrument-focused rehabilitation. Hand
Clin 19 (2):vii, 287–301.
Wiesendanger, M., A. P. Baader, and O. Kazennikov. 2006. Fingering and bowing in violin-
ists: A motor control approach. In Music, motor control and the brain, ed. E. Altenmuller,
M. Wiesendanger, and J. Kesselring. Oxford: Oxford University Press.
Winspur, I. 1998a. Dupuytren’s contracture. In The musician’s hand: A clinical guide, ed.
I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
———. 1998b. Nerve compression syndromes. In The musician’s hand: A clinical guide, ed.
I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
———. 1998c. Osteoarthritis. In The musician’s hand: A clinical guide, ed. I. Winspur and
C. B. Wynn Parry. London: Martin Dunitz.
———. 1998d. Surgical indications, planning and technique. In The musician’s hand: A clinical
guide, ed. I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
———. 2003. Controversies surrounding “misuse,” “overuse,” and “repetition” in musicians.
Hand Clin 19 (2):vii–viii, 325–29.
Winspur, I., and C. B. Wynn Parry. 1997. The musician’s hand. J Hand Surg (Br) 22 (4):433–40.
———. 1998. The musician’s hand: A clinical guide. London: Martin Dunitz.
———. 2000. Musician’s hands: A surgeon’s perspective. Med Prob Perform Artists 15
(1):31.
Wolf, F. G., M. S. Keane, K. D. Brandt, and B. M. Hillberry. 1993. An investigation of finger
joint and tendon forces in experienced pianists. Med Prob Perform Artists 8:84–95.
Wristen, B. G., P. Jung, A. K. G. Wismer, and M. S. Hallbeck. 2006. Assessment of muscle
activity and joint angles in small-handed pianists: A pilot study on 7/8-sized keyboard versus
the full-sized keyboard. Med Prob Perform Artists 21 (1):3–9.
Wynn Parry, C. B. 1998a. The interface. In The musician’s hand: A clinical guide, ed. I. Win-
spur and C. B. Wynn Parry. London: Martin Dunitz.
———. 1998b. Misuse and overuse. In The musician’s hand: A clinical guide, ed. I. Winspur
and C. B. Wynn Parry. London: Martin Dunitz.
———. 1998c. The musician’s hand and arm pain. In The musician’s hand: A clinical guide,
ed. I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
———. 1998d. Rheumatoid arthritis. In The musician’s hand: A clinical guide, ed. I. Winspur
and C. B. Wynn Parry. London: Martin Dunitz.
———. 1998e. Specific conditions. In The musician’s hand: A clinical guide, ed. I. Winspur and
C. B. Wynn Parry. London: Martin Dunitz.
Wynn Parry, C. B., S. Fischer, J. Williams, B. Gregor-Smith, C. Grindea, and J. Kember. 1998.
The interface. In The musician’s hand: A clinical guide, ed. I. Winspur and C. B. Wynn
Parry. London: Martin Dunitz.
Zaza, C. 1998. Playing-related musculoskeletal disorders in musicians: A systematic review of
incidence and prevalence. CMAJ 158 (8):1019–25.
 Chapter Four

Breathing in Singing and Wind Playing

Even though we all breathe constantly without being aware of it and can also read-
ily bring this under conscious control, the events that take place during inspiration
(breathing in) and expiration (breathing out) are far from intuitive. The reasons for
this will become clear later on in this chapter, but the consequence for singers and
wind players is that they often have misconceptions about breath control, which are
reflected in the language they use to describe the respiratory mechanisms that they
believe underlie singing and wind playing.
This chapter does not seek to describe an optimum strategy for respiration in musi-
cians. Research into the physiological processes that are involved has much further to
go before it can provide comprehensive and objective guidance. In the meantime, it
must be left to singers, wind players, and their teachers to decide on the best approach,
which will be determined by each performer’s experience and aesthetic goals and, as
we shall see, may also be influenced by that individual’s particular body type. Instead,
this chapter has two main objectives. The first is to provide an accurate description
of the activity of respiratory muscles and the movements they produce during normal
respiration. This is an essential starting point for all singers and wind players and will
allow them to communicate and discuss aspects of their technique accurately and
unambiguously. The second objective is to review the results that have been obtained
from the scientific analysis of respiratory activity underlying singing and wind play-
ing. It will be clear from this section that the small body of knowledge that currently
exists is still quite rudimentary, not least because it is based on studies of only a few
individuals. Furthermore, in order to make such investigations practical, the number
of parameters measured in each study must be quite restricted. The musical tasks
the subjects are asked to carry out must also be both few and simple so that they are
readily repeatable and comparable between individuals and generate an amount of
data that can realistically be analyzed. The published studies of singers focus pre-
dominantly on those who are classically trained. To a large extent this is because the
respiratory demands made on them are greatest, as they perform without amplification
and may have to make themselves heard over a full orchestra. As we shall see, this has
a major effect on the nature of the sound they produce. Finally, of course, we cannot

101
102 Chapter Four

Figure 4-1. Lung structure. The trachea divides into two bronchi, one entering each lung. Like
the trachea, these are held open by rigid rings of cartilage. They divide into progressively smaller-
diameter airways called bronchioles, which lack cartilaginous rings. The narrowest of these are
the terminal bronchi, which end in alveoli. Here oxygen passes to the red blood corpuscles, while
carbon dioxide diffuses out of the blood plasma in the opposite direction. The lungs are covered
with a sheet of visceral pleura, while the inner surface of the chest is lined with a sheet of parietal
pleura. These are held in contact by the surface tension of a film of watery fluid that prevents
the lungs from collapsing.

hear the voices of the participants in these studies and so we cannot assess whether we
find the outcome of the different approaches used to be to our particular tastes. Despite
these limitations, the information that is currently available may at least provide some
food for thought and a basis for personal experimentation.

THE STRUCTURE OF THE LUNGS

During inspiration, air passes through the larynx into the trachea (windpipe), which
divides into two tubes called bronchi, one of which enters each lung (fig. 4-1). The
trachea and the bronchi and their major branches in the lungs are held open by a se-
ries of rings made of cartilage. Once inside the lung, the bronchi continue to divide
repeatedly into smaller and thinner-walled tubes, ending in fine branches known as
respiratory bronchioles. These lack any cartilaginous scaffolding and terminate in
large numbers of small sacs called alveoli, where the air is brought into close contact
with very narrow, thin-walled blood vessels known as capillaries. Here, oxygen and
carbon dioxide need travel only a short distance as they diffuse into and out of the
blood, making the process very efficient. Oxygen diffuses from the alveolar air into
the capillaries and binds to the hemoglobin contained within the red blood cells. Car-
bon dioxide produced by the tissues of the body dissolves in the blood plasma, the
watery fluid that carries the red blood cells, and on reaching the lungs diffuses out in
the opposite direction into the alveoli. The structure of the lungs is designed not only
 Breathing in Singing and Wind Playing 103

Figure 4-2. Diagram showing lung function parameters. During normal quiet breathing only the
tidal volume is used. With a maximal intake of breath the inspiratory reserve volume is used. If
the breath is then driven out completely, we pass through the tidal volume band into the expira-
tory reserve volume. This exhalation uses all of the vital capacity. This leaves the residual volume
which cannot be expelled because the chest has a rigid outer wall on three sides. As the outer
surface of the lungs is held in contact with this, they cannot collapse completely.

to bring the respiratory and blood systems very close together, but also to pack within
a small volume an enormous surface area for gaseous exchange. There are about 300
million alveoli in each lung, and their total surface area is approximately 100 square
meters. The composition of air taken into the lungs is approximately 21 percent oxy-
gen and 79 percent nitrogen. The nitrogen is inert and plays no role in respiration.
In exhaled air the oxygen concentration falls to 17 percent, while the carbon dioxide
that has diffused out of the blood makes up 4 percent. If we hold our breath, it is the
buildup of carbon dioxide in the bloodstream that is the stimulus that ultimately forces
us to breathe again. During singing or wind playing, the lungs are used not only for
respiration but also as a reservoir for the air that is used to set up the vibration of the
larynx, lips, or reed. The limitation to the length of the breath is generally determined
by how quickly the available volume of air in the lungs flows out. In some instruments
such as the oboe, however, the flow rate is very slow and so the buildup of carbon di-
oxide can become the limiting factor, requiring the remaining air to be quickly exhaled
before the air in the lungs is exhausted, and a new breath is taken. The amount of car-
bon dioxide dissolved in the blood affects its acidity (or pH), which in turn influences
104 Chapter Four

its other properties, such as the ability of the hemoglobin to bind oxygen. If we breathe
very rapidly when not physically exerting ourselves, then too much carbon dioxide
may be lost from the plasma. This can be caused by performance anxiety (Widmer et
al. 1997) or an unsuitable breathing strategy. One of the immediate consequences of
this may be dizziness due to a constriction of blood vessels in the brain and an increase
in heart rate. Consequently it is important not to overbreathe when using the lungs as
bellows during singing or wind playing.

Muscle Action and Skeletal Movement during Normal Breathing

During normal quiet breathing (i.e., when we are sitting passively and neither speak-
ing nor carrying out any physical activity) only about one-tenth of our total lung
capacity is used (fig. 4-2). This is called the tidal volume and is typically of the order
of about 500 milliliters. Tidal breathing uses only the muscles of the chest, and each
breath ends when all of the respiratory muscles are relaxed. At the end of a quiet
breath, more air (the expiratory reserve volume) can be expelled using the muscles
of the abdominal wall. If we fill our lungs as full as possible, then blow out all the
air that we can, the volume exhaled constitutes the vital capacity of the lungs. About
1.5 liters of air remains in the lungs because the rib cage cannot collapse completely;
this is known as the residual volume. The residual volume plus the expiratory reserve
volume make up the functional residual capacity. The total lung capacity is the vital
capacity plus the residual volume. Singing and wind playing require much larger
volumes of air than we need for quiet breathing and for most wind instruments this
must be delivered at much higher pressures. For both activities, we often take breaths
that use a large percentage of our vital capacity and end well below the level of our
functional residual capacity. Though it is intuitively obvious that vital capacity and
the pressure that can be achieved will vary with body size, the extent of normal range
may come as a surprise. From our own observations of a group of college wind play-
ers aged between eighteen and twenty-one and in good health, it is clear that a woman
with a height of five feet two inches (157 cm) may have a vital capacity of only 3.5
liters, while that of a six-foot two-inch (188 cm) man may approach 7 liters! Clearly
this needs to be fully appreciated by singing and instrumental teachers.
Respiration involves the action of muscles in both the chest (the thorax) and the
abdomen, the two large chambers that form the trunk. The thorax contains the lungs
and heart, and its side walls are formed by the rib cage. Beneath the thorax lies the
abdomen, whose walls are muscular and which contains the liver, gut, and other major
visceral organs. The boundary between the thorax and abdomen is formed by a hori-
zontal sheet of muscle called the diaphragm (figs. 4-1, 4-3, 4-4). Though this plays a
key role in respiration, its action is often misunderstood.
During inspiration, expansion of the lungs results in a transient reduction in their
internal pressure. As a consequence, air from the atmosphere flows in through the
trachea until the pressure inside and outside the lungs is equalized. During expiration,
the lungs are compressed, increasing their internal pressure and airflows out until the
internal and external pressures are once more equal. Lung tissue itself contains no
muscle, so changes in lung volume are brought about indirectly by the movement of
 Breathing in Singing and Wind Playing 105

the walls of the thorax. Each lung is covered by a thin, transparent membrane that
forms a sac (the visceral pleura) that is attached to its surface (fig. 4-1). The inside
of the thorax is also lined by a similar sac (the parietal pleura). These two layers of
pleura lie in contact, and during respiratory movements they slide smoothly over each
other. They can do this because they are lubricated by a thin film of fluid that lies
between them. If the two layers briefly stick to each other, as sometimes happens with
lung infections such as pleurisy, then a sharp pain is experienced in the chest when a
breath is taken in. The layer of fluid also holds the two layers of pleura in contact by
surface tension. This is the same principle we use to stick a plastic toll pass holder to
a car windshield. If the two layers of pleura did not adhere in this way, then expand-
ing the chest wall would not expand the lungs. If the chest wall is punctured and air
gets into the space between the visceral and parietal pleura, the lung (which is elastic
and is stretched when in contact with the chest wall) collapses away from the ribs and
becomes useless for respiration. This is extremely dangerous as the lung cannot be
inflated again until the air between the two layers of pleura is removed.
Normal respiratory movements mainly involve the muscles of the thorax and of the
abdomen (figs. 4-3, 4-4). The lateral walls of the thorax are formed by the ribs, which
are hinged from the spine (vertebral column) at the back and are attached by stiff car-
tilage to a flat bone called the sternum (breastbone) at the front. Several sets of muscle
move the ribs during respiration. Between adjacent ribs run the intercostal muscles
(fig. 4-3), of which there are three sheets in each rib space; the outermost sheet is
formed by the external intercostals, whose main action is to swing the ribs upward and
outward when air is drawn into the lungs. The inner two layers (the intermediate and
internal intercostals) pull the ribs downward to force air out of the lungs. The floor

Figure 4-3. The major respiratory muscles of the thoracic and abdominal walls. The external
intercostal muscles that run between the ribs are muscles of inhalation. A. The lateral abdominal
wall muscles and the rectus abdominus muscle are involved in exhalation. There are three lay-
ers of lateral abdominal muscles whose fibers run in different directions (indicated by the lines).
B. The diaphragm forms the floor of the chest cavity and is attached to the liver. C. Contraction of
the lateral abdominal walls during exhalation compresses the viscera and pushes the diaphragm
upward.
106 Chapter Four

of the thorax is formed by the diaphragm, which is a sheet of muscle covered with
parietal pleura. The liver is attached to the underside of the diaphragm, and when the
diaphragm is at rest, the liver pushes it up into the thorax as a dome (figs. 4-3, 4-4).
The muscles of the abdominal wall also play a major role in respiration. The anterior
wall of the abdomen is formed by the rectus abdominus. This is a large, vertically run-
ning muscle divided into segments that when well developed are popularly known as
a “six-pack.” The side walls of the abdomen are made up of three overlapping layers,
the internal and external obliques and the transverse abdominal muscles (we will refer
to these as the lateral abdominal muscles). Though each sheet is thin, their fibers run
in different directions so that together they form a strong muscular wall. The develop-
ment of these muscles is a major goal of the Pilates exercise system. Several other
muscle groups are also involved in respiration; however, the ones described above are
the most important, and the mechanics of respiration can be readily understood on the
basis of their actions. We will therefore consider how these operate first.
The abdomen is effectively a sealed chamber, while the space in the chest that con-
tains the lungs is open to the outside via the trachea and mouth. The chest and abdomen
are separated by the muscular diaphragm, which at rest, as we have seen, is pushed
upward into the chest by the abdominal organs. During inspiration, the diaphragm
contracts and therefore flattens, causing the liver to be pushed downward like a piston
(fig. 4-4). At the same time, the external intercostal muscles raise the chest. The joints
between the ribs and the spine allow the ribs to swing outward and upward, which
increases both the width of the thorax and the distance between the vertebral column
and the sternum, as well as lifting the rib cage upward. The combined action of the
diaphragm and external intercostal muscles is to increase the volume of the thorax and
reduce the pressure of the air within the lungs. Air then flows in through the mouth and
nose, until the pressures inside and out are equalized. When we breathe very energeti-
cally (for example after vigorous exercise), movements of the thorax may be aided by
muscles in the neck that attach to the upper two ribs or to the collarbone (clavicle), but
under normal circumstances, these muscles are not much used.
The abdomen is supported at the back by the incompressible bony pillar formed
by the lumbar spine, but its lateral and anterior walls are purely muscular, so when
the diaphragm contracts and descends, these walls must bulge outward. The extent
and effectiveness of the movements of the diaphragm are influenced by posture. If a
sitting player or singer is slumped forward, the contents of the abdomen are forced
farther upward into the thorax than when standing. The walls of the abdomen are also
less free to expand, and a more forceful action of the diaphragm is necessary for full
inspiration. This will force the body into a more upright position. If the performer is
reclining, the abdominal organs slump back against the spine when the diaphragm
descends. The muscular wall of the abdomen does not bulge outward and the rib cage
does not fall so far passively during expiration. In both situations the result is a reduc-
tion in vital capacity—the amount of air available for singing or wind playing.
To expand the thorax, the inspiratory muscles have to overcome not only the effect
of gravity in raising the ribs, but also the natural elasticity of the lungs and the rib
cage. When the intercostal muscles relax during expiration, these forces cause the rib
cage to fall passively toward the rest position. The diaphragm relaxes and the muscles
Figure 4-4. A summary of respiratory movements and the action of the diaphragm.
108 Chapter Four

of the abdomen wall contract. Because the abdomen is a sealed chamber, compressing
its muscular walls forces it to elongate. There is no scope for extension at its lower
end, which is blocked off by the floor of the pelvis, so the contents of the abdomen
push the relaxed diaphragm upward into the thorax, again like a piston (fig. 4-4). The
contrasting roles of the diaphragm and the abdominal muscles are explained in anima-
tion 4-1 and shown running continuously in animation 4-2. Because they are the main
active component of expiration, the abdominal muscles make an important contribu-
tion to controlling the breath during singing. In addition, they are the primary means
by which the high air pressures needed to initiate and maintain the vibrations of the
reed or the lips in wind players are generated. The pattern of activation in the main
inspiratory and expiratory muscles is shown schematically in figure 4-5. This shows
a preparatory breath followed by a breath-supporting vocalization and ends with a
second silent breath. The breath pressure sustained during singing, which must induce
the vibration of the vocal folds of the larynx, is considerably higher than the pressure
generated in a silent breath.
The activity of respiratory muscles during singing and wind playing is more com-
plicated than has been described above. This is because of the requirement for very
precise control of the pressure and flow rate of exhaled air to regulate the volume,
pitch, and temporal structure of the notes. If we fill our lungs to capacity and exhale
passively using just the elastic forces of the thorax, the result is a sigh, which empties
them in a few seconds. The effect of the elastic forces on breath pressure is indicated
by what is labeled the relaxation line in figure 4-5. In order to prolong the breath or
maintain a constant breath pressure when the mouth is open, we have to resist the
elastic recoil of the thorax when it is high, at the beginning of the breath. This is called
respiratory braking and explains why the intercostal muscles remain active after the
onset of expiration during singing (fig. 4-5). With the descent of the ribs thus con-
trolled, we can further reduce or stop the flow of air by contracting the diaphragm.
However, in some singers, this may only be used for specific vocal effects of short
duration (see below). Later during the exhalation, the elastic forces fall below the level
necessary to maintain the required breath pressure and so the expiratory muscles of
the abdomen and the inner intercostal muscles are recruited. The lateral abdominal
muscles and the inner intercostals contract first. The rectus abdominus muscle makes
its contribution only toward the end of the breath to force the last of the air from the
lungs should this be necessary. As the rectus muscle runs from the rib cage to the front
of the pelvis, strong contractions pull the trunk forward and so this muscle is rarely
contracted in singing or wind playing.

Auxiliary Respiratory Muscles

In addition to the diaphragm, the intercostal muscles, and the muscles of the abdomi-
nal wall, a number of other muscles have been implicated in respiration, at least under
certain circumstances. These are known as auxiliary respiratory muscles and some
among them feature prominently in certain schools of vocal pedagogy. Most of the
muscles discussed in this section are shown in figure 4-6. Some act directly on the
ribs but others exert their effects indirectly via the shoulder girdle or upper arm. The
 Breathing in Singing and Wind Playing 109

Figure 4-5. Respiratory muscle activity during singing or speaking. This diagram demonstrates in
an idealized format the pattern of activity of the major respiratory muscles underlying an episode
of vocalization. The changing level of activity of the muscles is indicated by the slopes of the
triangles (Ladefoged, 1967).

principal roles of most of these muscles are quite unrelated to respiration, and some
may only be able to make a significant contribution to breath control when certain
postures are adopted or when the arm or shoulder blade is held fixed either by the ac-
tion of other muscles or by leaning the arms on a stationary surface.
The first set of auxiliary respiratory muscles we will consider are associated with
the trunk. These are the serratus posterior muscles, the latissumus dorsi, the pectoral
muscles, and the subclavius. There are two groups of serratus posterior muscles.
The serratus posterior superior muscles typically run obliquely downward from the
last vertebra of the neck and the upper thoracic vertebrae and attach to the second to
fourth ribs, which they are said to draw upward during forced inspiration. The serratus
posterior inferior muscles run from the last thoracic and upper lumbar vertebrae to the
lowest four ribs, which they pull downward. Considerable controversy has surrounded
the possible role of both sets of muscles in breathing. A comprehensive review of the
110 Chapter Four

Figure 4-6. The muscles shown are suggested to play an accessory role in respiration, though
they are best known for other unrelated actions.

available literature (Vilensky et al. 2001) suggests that these muscles probably have
no significant respiratory role but that the sensations generated when they are stretched
may influence the activity of muscles that are actively involved in breathing. The ser-
ratus posterior inferior muscles might, for example, contribute to the sensations in the
lower back during inspiration that are referred to by some singing teachers.
Some singing schools make considerable play of the role of the latissimus dorsi
muscle in breathing (Miller 1997). On the face of it, this is rather surprising as the
main function of this muscle is to move the arm. At one end it is attached to the lower
thoracic vertebrae and, indirectly through a broad, tough sheet of tissue, to the lumbar
vertebrae and the hip girdle. At the other end it attaches to the humerus (the bone of
the upper arm) just below the shoulder and thus forms the back wall of the armpit. Its
main action is to pull the arm downward. As a consequence, it is very well developed
in swimmers, who use it to pull themselves through the water. The suggestion that
it plays a role in respiration arises because of its attachment to the lower three ribs,
which it might conceivably raise during inhalation; however, this would only be effec-
tive if the upper arms were held rigidly immobile, which is not normally the case in
singers. Physiological studies of the muscle (Orozco-Levi et al. 1995) show it can be
used to aid respiration but usually only when the breath is drawn in through the mouth
against considerable resistance (i.e., the subject has to suck hard to inhale). Conse-
quently, it is highly unlikely that this muscle plays any significant role in breathing
movements underlying singing. The other possibility is that it helps maintain a posture
that is optimal for the supported voice; however, this has yet to be investigated.
Campbell (1970) considers that the pectoral muscles (pectoralis major and minor)
may also be brought into play during forced inhalation but not in normal breathing.
 Breathing in Singing and Wind Playing 111

The pectoralis major is attached at one end to the breastbone and the cartilages that
link this to the ribs and to the adjacent part of the collarbone. From these regions it
runs to the humerus of the upper arm. Its main action is therefore on the arm, but if this
is held fixed it will pull the upper ribs outward and upward. The pectoralis minor runs
from the third to fifth ribs to a spur (the coracoid process) that projects forward from
the shoulder blade. Its action is usually described as stabilizing the shoulder blade
or pulling it downward, but if the shoulder blade is held fixed by other muscles, the
ribs would be pulled upward; however, this muscle’s role in respiration has not been
investigated in any detail. The subclavius is a small muscle that holds the collarbone
against the junction of the first rib with the breastbone. It, together with the pectoral
muscles and the trapezius (see below) might be brought into play in so-called clavicu-
lar breathing (Miller 1997), in which the chest is raised as a consequence of lifting the
shoulders; however, this strategy is frowned upon in most schools of singing and is
likely to have a number of consequences that do not favor good vocal performance.
Without the engagement of these muscles, the shoulders can be readily lifted without
raising the chest or having any significant effect on breathing.
Many of the muscles of the neck are also multifunctional; their involvement in
breathing, though important for singers, is often viewed as a secondary role by anato-
mists. The primary function of muscles with an accessory respiratory function is to
stabilize or rotate the head (sternocleidomastoid muscle), tilt the head from side to
side (the scalene muscles), or raise the shoulders or pull them toward the spine (trape-
zius). The neck also contains other muscle groups that do not play any obvious role in
respiration. Singers often experience neck pain or tension that probably results from
stress and can interfere with performance (Pettersen and Westgaard 2005), and so
understanding the action of these muscles and reducing their activity where possible
is important in vocal teaching.
The actions of the trapezius muscle are complex. Though their most commonly
cited action is the raising of the shoulders in a shrugging movement, this is brought
about only by the upper part of the muscle, whose fibers run upward to the cervical
spine and the underside of the skull. The fibers of the middle part run horizontally
and draw the shoulder blades toward the spine, while the fibers of the lower part run
obliquely downward and pull the shoulders in that direction. Only the upper part is
generally referred to in the context of singing. In standard texts dealing with respira-
tion, it is generally said that the trapezius can aid inspiration by raising the shoulder
girdle. As the shoulder is attached to the thorax by the muscles described previously,
this pulls the rib cage upward (Campbell 1970). In singing, however, the trapezius
appears to be involved in expiration, during which it may compress the thorax by
pulling on the shoulders and rotating the shoulder blades toward the spine (Pettersen
and Westgaard 2004). The degree to which it is active varies greatly among sing-
ers, but tension in the muscle is said to be greatest among those who are relatively
inexperienced. It is well established that contraction of the trapezius is often linked
to emotional stress, and under these circumstances it may serve no physiological
function (Waersted, Bjorklund, and Westgaard 1994). Reducing shoulder tension is
generally regarded as beneficial to voice production. When singers are trained to do
so through biofeedback techniques, no compensatory changes are seen in the activity
112 Chapter Four

of other respiratory muscles, suggesting that the activity of the trapezius is superflu-
ous to most vocal tasks (Pettersen and Westgaard 2002, 2004).
As activity patterns of the sternocleidomastoid and scalene muscles during singing
have many similarities, we will deal with them together. One sternocleidomastoid
muscle runs on each side of the neck, from the skull just behind the ear to the breast-
bone (sternum) and the adjacent part of the collarbone (clavicle). If the muscle con-
tracts only on one side, this turns the head upward and to the opposite side, but when
both sides act together this thrusts the head forward. If the head is held in a fixed po-
sition, the two muscles pull the sternum upward, which helps draw air into the lungs.
There are three scalene muscles on each side of the neck. They run from the sides of
the neck vertebrae to the first and second ribs. If the neck is free to move, the scalenes
bend the neck to the side they are on, but if the neck is held straight by other muscles,
the combined action of the right and left scalenes is to pull the ribs they attach to
upward, aiding inhalation. Interpreting the activity of the sternocleidomastoid and
scalene muscles during singing is complicated by the fact that they are involved not
only in breathing, but also in positioning and stabilizing the head, which in turn has an
effect on the height of the larynx. The sternocleidomastoid muscles are generally not
involved in quiet breathing not related to singing, but there is often activity in the sca-
lenes. With deeper breaths they are progressively brought into play, and activity in the
scalenes increases (Campbell 1970; Pettersen and Westgaard 2005). As singers often
take breaths close to 100 percent vital capacity, it is common to find marked activ-
ity in both sets of muscles, particularly toward the end of inhalation. There is often a
brief peak of activation that coincides with start of phonation and more irregular bursts
of activity during sung phrases, which may reflect a role in resisting (and therefore
giving greater control over) expiratory movements (Campbell 1970; Pettersen and
Westgaard 2005). This activity was particularly marked in a study in which perform-
ers were asked to sing glissandi that rose to the uppermost part of the range and then
fell again (Pettersen et al. 2005). In most of the singers studied, the lowest part of the
thorax contracted first, while the upper part was kept expanded until the highest tone
was reached. A burst of activity then occurred in the sternocleidomastoid and scalene
muscles at the transition point where the upper part of the thorax began to fall. Pat-
terns of muscle activity during inhalation are nevertheless idiosyncratic, and some
singers appear to make little use of these muscles under most circumstances. This
may be related to the overall respiratory strategy used; for example, in one singer who
showed no peak of activity at the turning point of the glissando, the upper and lower
parts of the thorax fell in parallel during its entirety. More will be said later about the
variability in respiratory patterns exhibited by different individuals.
Among some singing teachers of the German school, contraction of the muscles of
the buttocks and pelvic floor is sometimes invoked as a component of breath support
(Miller 1997). It is, however, difficult to justify this on anatomical grounds. The pri-
mary function of the largest muscle of the buttock, the gluteus maximus, is to stabilize
or extend the thigh (draw it backward, e.g., during walking or running) and to rotate it
outward. However, in standing, when the foot is planted firmly on the ground, it can
be used to rotate the pelvis backward and hence contribute to holding the trunk erect.
Clearly this erect posture will contribute to efficient breathing, though the gluteus
 Breathing in Singing and Wind Playing 113

maximus is only one of the muscles responsible and makes only a relatively small
contribution to this. The muscles of the pelvic floor, the most prominent of which is
the levator ani, block off the lower opening of the abdomen and also contribute to the
sphincters (or seals) of the anus, urethra (the channel for urine), and vagina. Singers
may be encouraged to tense this muscle while clenching the buttocks during pas-
sages or phrases that require strong breath support by requests to “sit on the breath”
or “squeeze the dime” (Miller 1997). There is, however, no anatomical rationale to
explain why this might produce a significant benefit for singers. The muscles of the
pelvic floor can be weakened by damage during childbirth or as a consequence of
age, and as a result a sudden increase in abdominal pressure due to sneezing or even
laughter may cause leakage of urine (stress incontinence). It is possible that this has
led to the misguided notion that clenching the buttocks and tensing the pelvic floor is
needed to support the breath; however, the pressures produced in the chest and abdo-
men in the explosive action of sneezing far exceed those generated during singing.
Any signs of stress incontinence in singers should first be addressed by exercises de-
signed to strengthen the pelvic floor, which should obviate any requirement to clench
during performance.

The Concept of Support

The control of the diaphragm and intercostal muscles in combination with the contrac-
tion of the expiratory muscles of the abdominal wall constitutes a significant part of
what is usually called “support.” However, also embodied in this concept are greater
voice resonance, clarity, and range, which arise partly from other manipulations of the
vocal tract. Despite its importance, there is no single and unambiguous definition of
support, and the word has different shades of meaning for different singers and teach-
ers (Griffin et al. 1995). The terms Atemstütze and appoggio refer approximately to
the same phenomenon. Distinctions are made in some schools between singing with
“no support,” “chest support,” or “diaphragm support” (Winckel 1952); however,
these concepts are based on subjective sensations that can be misleading and may
reflect an incomplete understanding of the mechanics and anatomy of respiration
(Watson and Hixon 1985). In addition, it is almost impossible to assess subjectively
all that is going on physically during singing, not only because many muscles in the
vocal tract and respiratory system are being activated simultaneously, but also because
some key muscles (most crucially for the present discussion, the diaphragm) have
a relatively poor sensory nerve supply, meaning that we are not generally aware of
when they are active. There is, however, little disagreement on the beneficial effects
of support to the quality of the voice and a high degree of consistency in the recogni-
tion of supported singing among those professionally involved with singers (Sand and
Sundberg 2005).
In some schools of singing, the role of the abdominal muscles in support is empha-
sized (Thorpe et al. 2001), but it is only recently that objective analyses of the activity
of different muscle groups during singing and wind playing have been attempted.
Even in studies with this aim, the actions of the muscles are often inferred only indi-
rectly from measurements of thoracic or abdominal dimensions or pressure levels in
114 Chapter Four

different parts of the vocal tract. This may be due to the understandable reluctance
of experimenters and their subjects to use invasive techniques such as placing elec-
trodes directly into the muscles themselves. In its most general sense, support is the
regulation of air pressure and the velocity of airflow in the respiratory system during
expiratory episodes underlying singing or wind playing and is generated by a dynamic
interaction between expiratory and inspiratory muscles. Changes in air pressure below
the glottis (the opening between the vocal folds of the larynx) are largely driven by the
movement of the contents of the abdomen (i.e., the visceral organs) as they are pushed
up by the abdominal muscles. Their mass gives them considerable inertia, and obtain-
ing precise control of their movement requires, as we have seen, the concerted actions
of both the expiratory muscles of the abdominal wall and the inspiratory muscles of
the diaphragm and thorax. At certain stages of the respiratory cycle these may even
be active simultaneously, the strength of contraction of the antagonistic muscles being
continually adjusted to produce precise movements in the abdominal mass and control
changes in subglottic pressure (Leanderson, Sundberg, and von Euler 1987).
The concept of support is rather different for wind players, and its precise nature
depends to a large extent on the type of instrument. In the flute and recorder, the con-
trol issues most closely resemble those discussed for singers, as the mouth is either
fully open or the resistance to airflow very low. For these instruments, sound can be
generated using elastic recoil forces alone when the lungs are fully inflated (see be-
low). In reed and brass instruments, where the air column in the pharynx is stopped at
the embouchure, the pressure required to produce the sound is much greater than can
be generated by elastic recoil and always requires the involvement of the abdominal
muscles. For players of these instruments the word “support” will therefore refer
exclusively to the action of the expiratory muscles. As the abdominal and intercostal
muscles contain many sensory nerve endings, we are well aware when they are active.
The lack of sensation from the diaphragm may partly explain why breath support is
often inappropriately linked to this muscle. The word “diaphragm” is often used when
what is actually meant are the muscles of the abdominal walls, which, of course, have
the opposite action.

MUSCLE ACTIVITY DURING SINGING OR WIND PLAYING

It is not possible to obtain a detailed understanding of how the chest and abdominal walls
move during breathing by following their movements visually or by relying entirely on
subjective sensation. This information can be obtained, however, with a device known
as a plethysmograph. There are many designs of plethysmographs, but one of the most
straightforward to use (the inductive plethysmograph) employs elastic bands that are
placed around the chest and abdomen to monitor their movements. The bands contain
open coils of wire whose resistance changes as they expand and contract, providing a
measure of the changes in their circumference. The bands are light and comfortable and
do not interfere with the movements they are measuring. Demonstrations of how these
can be used to analyze singing and wind playing are shown in videos 4-1 and 4-2, and a
number of the studies referred to below use variants of this technique. The traces gener-
 Breathing in Singing and Wind Playing 115

ated have a sawtooth contour, with a rapid upward movement representing inhalation
and a more gentle slope seen during the exhalation that underlies singing or playing.
The most useful information that can be drawn from plethysmography is provided by
the angles of the slopes of the traces from the bands on the chest and abdomen and their
relationship in time. This is difficult to assess visually; however, the chest movement
can be plotted against the abdominal movement to create what is known as a Konno-
Mead diagram (fig. 4-7), which is relatively simple to interpret intuitively (Konno and
Mead 1967), even for those with no background in respiratory physiology. The practi-
cal application of Konno-Mead diagrams to breathing in singers and wind players is
demonstrated in videos 4-3 and 4-4. In the idealized examples shown in figure 4-7, the
movement of the abdomen is represented along the horizontal axis and that of the tho-
rax on the vertical axis. Two types of reference lines are often shown on the traces. The
first are isovolume lines, which are obtained from inspiratory and expiratory move-
ments made with the mouth and glottis closed. Any slope on the diagram that is parallel
to these lines represents complementary movements of the chest and abdominal wall
that result in no overall change in lung volume. The other reference is the relaxation
line, which represents the passive falling of the chest and recoil of the abdomen in the
absence of any muscular activity, for example, during a sigh. It is important to note that
the graph has no time dimension, so the duration of each phase of respiration cannot
be determined from these diagrams. You will see from the section of the videos show-
ing the lines being generated in real time that some segments of the traces are drawn
rapidly (e.g., during inhalation) and others more slowly. Konno-Mead diagrams have
been quite extensively used for studying respiration in singers, but so far appear in only
one paper on wind playing (Cossette, Sliwinski, and Macklem 2000). They have great
potential for providing an objective analysis of breathing patterns between singers of
different types and players of the same or different instruments, and we are currently
performing extensive studies using this technique at the Royal Welsh College of Music
and Drama and at Cardiff University.
If we return to an examination of the idealized Konno-Mead diagrams in figure 4-7,
we can see that at the beginning of an inspiration, the contraction of the diaphragm
pushes the contents of the abdomen downward, causing its wall to bulge outward. This
may induce a brief decline in thoracic volume as the contraction of the diaphragm
pulls the lower edge of the thorax inward, followed by the engagement of the inter-
costal muscles that raise the ribcage, rotating the ribs outward, while the diaphragm
continues to contract. At the transition between inspiration and expiration, the ab-
dominal muscles contract, which initially pushes the chest wall farther out. This does
not necessarily involve a change in lung volume, as it may run nearly parallel to the
isovolume line; however, for loud singing or speech, some reduction in volume is usu-
ally present. The movement represents what is called “prephonatory posturing,” which
may optimize the position of the chest wall for the start of vocalization, initiating the
support mechanism and raising lung pressure sufficiently to drive the vocal cords, lips,
or reed at a level sufficient to generate the note to be produced. Expiration then be-
gins with a continued contraction of the muscles of the abdominal wall together with
the downward rotation of the ribs driven by the elastic forces released as the external
intercostal muscles relax.
116 Chapter Four

Figure 4-7

Three different strategies for expiration can be distinguished using Konno-Mead

diagrams. In the first, the abdomen and the chest move inward proportionately at the
same time, resulting in an almost straight diagonal line running downward and to the
left (fig. 4-7A). In the second, the chest falls first and the abdominal wall only moves
inward later (fig. 4-7B), while in the third, the order is reversed so that abdominal
movement precedes that of the chest (fig. 4-7C).
 Breathing in Singing and Wind Playing 117

Figure 4-7. Konno-Mead diagrams can be used to illustrate the changing dimensions of the chest
and abdomen during breathing. Idealized diagrams for three different breathing strategies are
shown here. The vertical axis represents the horizontal depth or circumference of the chest (de-
pending on method) and the horizontal axis, the same dimension for the abdomen. The oblique
parallel lines are isovolume contours; that is, movements parallel to these lines produce no over-
all change in lung volume. The relaxation line represents the change in the dimensions when the
chest wall is allowed to fall passively. During quiet breathing the movements of the chest and
abdomen are very small and lie close to the relaxation line. During singing, the movements are
much larger. At the start of exhalation (a) there is initially no change in chest volume; the line
between (a) and (b) runs parallel to the isovolume line. This is called prephonatory posturing
and is required to increase the air pressure in the lungs before singing. The voice is produced
from (b) onward. The first phase of inspiration is mainly due to the outward movement of the
abdominal wall as its muscles relax and the diaphragm is pushed downward, but the contribu-
tion of the chest progressively increases as the external intercostal muscles pull the ribs upward
and outward. Note that the Konno-Mead diagram has no time dimension, so it is not possible to
deduce from it how long each phase of the cycle lasts. A. Comparison of the changing dimensions
of the chest and abdomen seen during quiet breathing with that during singing. During the vocal
phase of exhalation, the line runs almost diagonally, indicating that the walls of the chest and
abdomen are moving inward at the same time. The same phase in B is at first entirely due to an
inward movement of the chest (i.e., the trace falls vertically); when this ends, contraction of the
abdominal muscles takes over. In C, the opposite occurs: the first part of the exhalation is entirely
due to abdominal muscle activity (indicated by the horizontal direction of the trace) but then
switches to chest movement. Konno-Mead diagrams are further explained in videos 4-3 and 4-4.

Singing
In objective studies of respiration in singers, one feature emerges strongly, namely,
that patterns of chest and abdominal movements and of muscular activity vary widely
between individuals. This may in part reflect different schools of singing pedagogy,
but there are clearly other factors at work as well. In some singers, the changes in tho-
racic and abdominal dimensions run in parallel as in figure 4-7A (Thorpe et al. 2001),
while others are taught to hold the rib cage out as the muscles of the abdominal wall
contract (fig. 4-7C). The dramatic effects of different strategies can be seen in video
4-3. The variability in relative contribution of chest and abdominal movements may
also reflect differences in posture or the biomechanical properties of the thorax and
abdomen in different individuals for whom one or another strategy may be more suited
(Wilder 1983; Hoit and Hixon 1986). Individuals can be classified into three broad
categories according to body type (somatotype): endomorphs, who are plump; ecto-
morphs, who are slight; and mesomorphs, who are muscular (fig. 4-8). A preliminary
study of tidal breathing and breathing during speech suggests that individuals with
different somatotypes may use quite different respiratory strategies (Hoit and Hixon
1986). Unfortunately, this work is based on data obtained from only twelve subjects.
The categorization of these three discrete somatotypes is now also outdated, given the
ease with which we can obtain information on a wide range of parameters reflecting
body composition. Given the potential importance of the concept that optimal breath-
ing strategies (and perhaps other aspects of technique) might be related to somatotype
in singers (and indeed in other performers), a proper analysis of this question should
be given a high priority.
118 Chapter Four

Figure 4-8. The differences in respiratory patterns between individuals with different body types
as proposed by Hoit and Hixon (1986).

If the conclusions of Hoit and Hixon are correct, it is likely that the more demanding
breathing strategies required by singing will show similar variability. They concluded
that endomorphs use longer and slower breaths during normal quiet breathing and
appear to employ predominantly abdominal mechanisms for breathing during speech
(fig. 4-8). This would have some similarities with what is sometimes called the “pear-
shaped up” breathing posture in singers (fig. 4-9), in which the abdomen is strongly
distended at the end of inhalation, while the chest is kept high for as long as possible
 Breathing in Singing and Wind Playing 119

Figure 4-9. Two breathing strategies sometimes described in singers. In the “pear-shape down”
posture, the chest is held high, and the aim is to exhale mainly through contraction of the abdomi-
nal muscles. When seen from the side, the cavities of the chest and abdomen have the shape of
an upside-down pear (see outline). The chin is held down, allowing the larynx to descend. In the
“pear-shape up” posture, the chest also falls during expiration. The singer leans back and raises
the chin, which tends to pull the larynx up. The consequences of these strategies for the voice
are discussed in chapter 5.

during the exhalation (Titze 2000). This should produce a Konno-Mead trace similar
to that in figure 4-7C. With this strategy the contraction of the abdominal muscles
during exhalation may cause a paradoxical upward movement of the chest wall (in the
direction normally seen during inhalation). The greater use of the abdominal muscles,
coupled with the longer breaths, pushes the diaphragm upward into the thorax farther
than during shorter, shallower breaths, or where inhalation is predominantly driven by
raising the chest wall. The result is a greater stretching of the diaphragm in its passive
state, which may make it more effective in increasing the volume of the thorax when it
contracts during inhalation. It is generally accepted that individuals develop the breath-
ing strategies that are energetically most cost effective for them, so the predominance
of abdominal mechanisms in endomorphs presumably suits those with large body mass.
This might be necessary because of the greater mass of the visceral organs (especially
the liver, which hangs from the underside of the diaphragm) and the more substantial
120 Chapter Four

oxygen demands of a larger body. The ectomorphs take shorter, shallower breaths
during normal quiet breathing and make much less use of abdominal volume changes
during speech breathing, in which most of the change in lung volume is brought
about by the rising and falling of the chest wall (fig. 4-7B). This would resemble the
“pear-shaped down” strategy of some singers, in which both chest and abdomen work
together (fig. 4-9). Mesomorphs use breathing strategies that lie somewhere between
those of the endomorphs and ectomorphs.
Despite the fact that many singers favor the “pear-shaped up” posture (which is
probably equivalent to the “noble” posture described by some singers), direct mea-
surements from performers suggest that changes in the dimensions of the chest wall
always make by far the greatest contribution to changes in lung volume (Thomasson
and Sundberg 2001b). In this study, the contribution of the abdominal wall varied
considerably between three male singers (all baritones), but the question of body type
or breathing strategy was not addressed. However, even if the movement of the ab-
dominal wall is relatively small, the degree of tension in the muscles that form it will
be a significant factor in ensuring that the fall in the chest wall drives air efficiently
through the larynx. What actually takes place during “pear-shaped up” and “pear-
shaped down” breathing and the possible advantages of the two strategies therefore
remain uncertain (Titze 2000). As will be seen in chapter 5, one significant factor may
be the effect that these postures have on laryngeal height.
As can be seen in figure 4-10, singers use most of their vital capacity when required
to sing long phrases while even for shorter ones, lung volumes decline well into the
zone of the functional residual capacity. However, the breath intake is matched to
phrase length so that excess air does not have to be exhaled at the end, which would be
a waste of both energy and time, delaying the onset of the following phrase and possi-
bly inducing the effects of hyperventilation. In quiet speech, lung volumes tend to stay
mainly above the line marking the functional residual capacity, while in declamatory
speech at the intensity used by actors in live theater, the pattern begins to resemble
that used in singing. One prominent feature of the lung volume traces for declamatory
speech and singing is a sharp decline immediately before the onset of the sound. This
represents prephonatory posturing, which is necessary to raise lung pressure to a level
appropriate to generate the vibration of the vocal folds for the succeeding note (see
videos 4-1 and 4-3).
Physiological studies of respiratory activity during singing reinforce the idea that
different singers employ quite distinct strategies to achieve the same goals (Griffin
et al. 1995; Thorpe et al. 2001), though the respiratory pattern in a given individual
asked to repeat the same exercise is highly consistent (Thomasson and Sundberg
2001a). This is nowhere more true than in the use of the diaphragm. In some sing-
ers it is active not only during inhalation, but remains so during the initial stages of
exhalation when lung volume is high. Its activity then declines progressively as the
lungs deflate during singing. By pushing the floor of the chest cavity downward, it
reduces the change in volume caused by the elastic recoil forces of the lungs and rib
cage. In other singers there is no evidence of any diaphragmatic activity during phona-
tion. These individuals presumably rely solely on the intercostal muscles to slow the
fall of the thorax. There are several factors that might support a greater role for the
Figure 4-10. Lung volumes used in speaking and singing the same text (the Welsh
hymn “Cwm Rhondda,” heard in video 4-2). In the upper part of each figure, changes
in lung volume obtained using plethysmography (expressed as a percentage of vital
capacity) are shown during the passage, which requires several breaths. During quiet
recitation, the respiratory trace remains above the functional residual capacity (FRC)
line, which represents the level at which all respiratory muscles are relaxed. With
declamatory speech at a volume appropriate for a stage performance, the trace dips
frequently below the FRC into the range where the abdominal muscles are engaged.
In supported singing using the projected voice, this extends almost to the lower limit
of vital capacity. As the sung phrases for this piece are relatively short, inhalations
do not approach 100 percent vital capacity. This does occur, however, when the
singer is asked to produce the longest note possible (bottom trace). In the traces for
declamatory speech and singing, each exhalation starts with a sharp reduction in lung
volume that precedes phonation (prephonatory posturing). This raises the air pres-
sure in the lungs to a level appropriate for the onset of the vocal fold vibration.
122 Chapter Four

intercostal muscles over the diaphragm in respiratory braking. First, the thoracic wall
has a much larger area of contact with the lungs (about 75 percent) compared to the
diaphragm, and so to produce the same pressure change, the movement of the ribs can
be much smaller than that required of the diaphragm. Second, the ribs are raised by
a series of short intercostal muscles (one set in each rib space) that will allow greater
control and range of contraction than the diaphragm, which is a single sheet of much
longer muscle fibers. Third, when the singer is upright, particularly when standing,
the viscera pull the diaphragm down passively anyway (Watson and Hixon 1985),
which may render significant contraction of the diaphragm unnecessary. Furthermore,
using the diaphragm in this way eats into vital capacity in a way that braking using
the intercostals does not. Finally, as previously mentioned, there is little subjective
sensation of diaphragm contraction, so its effects can only be appreciated indirectly
through tension changes in the thoracic or abdominal wall. This may reduce the level
of precision possible in its control.
Because of the controversy concerning the role of the diaphragm and the importance
attributed to it by many singing teachers, several studies have been designed to gain
a detailed objective assessment of how it operates during singing. One experiment, in
which a group of highly trained singers were asked to repeatedly sing the unvoiced
consonant “p” followed by the vowel “a” (pa, pa, pa, pa, pa, pa . . .), exemplified the
different patterns of diaphragmatic activity that are used by different individuals to
create an apparently similar effect (Leanderson, Sundberg, and von Euler 1987). In
one singer, the diaphragm remained inactive throughout the expiration. In a second,
there was a tonic contraction of the diaphragm to counter elastic recoil of the thorax
at the beginning of the breath. In a third, the diaphragm was active only phasically at
the beginning of the breath, during the vowel element, reducing subglottal pressure
transiently. The diaphragm then relaxed (allowing subglottal pressure to rise) during
the occlusion of the lips for the “p,” thus emphasising the explosive transition between
the “p” and the “a.” In a fourth singer, the activation of the diaphragm was similar
to the third, except that this time the peaks of its contraction coincided with the “p,”
which would decrease its explosive character but reduce breath expenditure. We are of
course left wondering whether the results of these four strategies (particularly the ones
involving phasic activity of the diaphragm) sound different to the listener, something
a written report cannot tell us, but as they appear to be properties of the individual
techniques of the singers concerned, they may have contributed to the individuality of
each singer’s vocal character. On the other hand, they might equally reflect a slightly
different interpretation of what was being asked for in the task.
Differences in strategy were again apparent when three of the same singers were
asked to sing alternating notes an octave apart (fig. 4-11). In two of the singers, the
diaphragm contracted to reduce subglottal pressure as the pitch dropped almost instan-
taneously from the high note to the low (fig. 4-11A). This effect was most marked at
the beginning of the breath, when thoracic recoil pressures were high. One subject,
however, who used his abdominal wall musculature to raise abdominal pressure to a
much higher level than the others, contracted the diaphragm most strongly during the
high notes (fig. 4-11B), presumably counteracting this with a considerable increase in
abdominal muscle activity! One explanation suggested for this behavior is that under
 Breathing in Singing and Wind Playing 123

Figure 4-11. Different ways of using the diaphragm during singing. The diagrams are idealized
representations of actual traces shown in Leanderson et al. (1987). Singers were asked to alter-
nate repeatedly between two notes an octave apart. Some singers use the strategy shown in A,
where the diaphragm contracts rapidly at the end of the high note to produce a rapid drop in
thoracic (lung) pressure. As the lungs gradually deflate during the breath, the elastic recoil forces
of the chest decline and so less diaphragmatic activity is needed. However, one singer used a dif-
ferent strategy (B) in which the diaphragm contracts during the high note. The thoracic pressure
generated is similar to that in A, but the singer here uses very high abdominal pressure and so
contraction of the diaphragm counteracts this to ensure that the thoracic pressure does not rise
excessively. This paradoxical use of the diaphragm is not unusual and has been shown in other
studies. The dashed traces in B represent the peak of abdominal pressure and the shape of the
diaphragm trace from A emphasizes how these differ.

these conditions the diaphragm compresses the contents of the abdomen and forces
it to expand outward, increasing the diameter of the base of the rib cage. This may
be used to keep the thorax expanded for as long as possible during the singing breath
(the pear-shaped up posture), which is encouraged by some teachers (Leanderson,
Sundberg, and von Euler 1987). Another possibility is that if the mass of the abdomen
was large, small complementary changes in two opposing forces might allow greater
control over its movement up into the chest cavity.
By contrast, in other situations all of the singers used the diaphragm in the same
way. The group was asked to perform a Renaissance trillo, in which the same note is
rapidly repeated interspersed with very brief silences. (The ornament heard, for ex-
ample, in the Monteverdi Vespers is sometimes unflatteringly described as the goat’s
trill!) In this rather extreme vocal task, the diaphragm was always activated phasically
to create the silences between the notes. The strategy used for singing coloratura pas-
sages was also consistent. Surprisingly, perhaps, there was little or no contribution of
the diaphragm here to the rapid fluctuations in subglottal pressure that accompany the
transitions between each note, which are presumably mainly derived from activity of
the muscles of the abdominal wall.
Given the lack of sensory feedback available from the diaphragm, the question of
whether it can be brought under conscious control at all is an important one. It has
124 Chapter Four

been shown that if diaphragmatic pressure is recorded electronically and presented to

singers on a computer screen, they can quickly learn to activate it at will (Leanderson,
Sundberg, and von Euler 1987). This generally has the effect of giving the voice a
“darker” or more “covered” quality, due to the suppression of the higher harmonic
partials of the note that apparently occurs as a result of a shift toward flow-phonation
from a more pressed state of the vocal folds (see chapter 5). However, it is also pos-
sible with practice to return to a “bright” sound even with the diaphragm active. This
implies that the change in tone is a secondary effect of diaphragmatic action, perhaps
due to greater tension on the vocal tract or to a change in posture. The experiment also
revealed that when the diaphragm is active, there is a reduction in the variability of the
formant frequencies of the notes produced. The reasons for this are currently unclear,
but it may be the result of a more stable positioning of the larynx.
One distinctive feature of the voice of most modern classical singers is vibrato.
Though there is some variation in its nature between individuals, it is typically a
fluctuation in the fundamental frequency of the note of about 50 cents at 6–8 hertz
(6–8 cycles per second). It appears to be associated with laryngeal muscle activity
(see chapter 5), particularly that of the cricothyroid muscles, which stretch the vocal
folds, increasing the pitch of the note. There also seems to be some involvement of the
vocalis muscles (which tense the vocal folds) and the lateral cricoarytenoid muscles
that bring them together, closing the glottis and lowering pitch (Sundberg 1999). Os-
cillations in phase with the vibrato may also be present in the muscles of the pharynx
and tongue, though it is not likely that these would make a significant contribution
to the sound. During the vibrato, fluctuations in subglottic pressure and airflow can
usually be measured, but their origin is unclear. Though these could be a secondary
effect of the oscillations in the vocal folds themselves, they could also result either
from rhythmic activity in the diaphragm (at high lung capacities) or in the muscles of
the abdomen. As will be seen below, all three of these elements may contribute to the
production of vibrato in the flute.

Breathing and Voice Projection

The audibility of a singer’s voice against the background of an orchestra is dependent
on what is known as the “singer’s formant” or “ring” (chapter 5). Though this is largely
due to a manipulation of the intrinsic resonances of the vocal tract, it may partly be a
consequence of the respiratory element of support. In a study of the supported voice
in four male and four female singers (Griffin et al. 1995), it was concluded that in the
males this was associated with an increase in subglottal pressure and peak airflow rate,
accompanied by a lowering of the larynx. However, the supported voice in females
(and particularly sopranos) was reported to employ a different strategy, being linked
to tighter closure of the vocal folds (i.e., leading to more pressed phonation) in the
upper range, while at lower pitches it actually involved a lower subglottal pressure.
Thorpe and colleagues also examined the relationship between respiratory activity and
vocal projection; their results partially supported the findings of Griffin et al. (1995).
The five singers who took part in the study had all been taught a strategy for voice
projection by the same teacher (Thorpe et al. 2001), which stressed the synchronized
 Breathing in Singing and Wind Playing 125

activation of the lateral abdominal muscles (internal and external obliques, transversus
abdominus) during phonation and their complete relaxation at the onset of inspiration.
The initial lung volumes used by the singers for each phrase (expressed as a percent-
age of vital capacity) were the same for projected and unprojected phrases despite
the fact that sound intensity increased, though they were of course smaller for much
quieter singing. This finding probably reflected the fact that because of their training,
the singers had already optimized their respiratory strategies and the initial lung vol-
ume was close to capacity. The lateral dimension of the thorax was increased during
projected singing while that of the abdomen was reduced. Both of these changes are
consistent with increased activity in the lateral abdominal muscles and therefore sug-
gest that the theory behind the strategy was being put into effect. Finally, there was a
reduction in the rate of airflow through the vocal tract in the projected voice, which
again suggests that this state is brought about by changes in the positioning and vibra-
tory patterns of the vocal folds so that they increase the resistance to the airflow.
While it is understandable that both studies were able to recruit only small num-
bers of singers, this raises some problems for their interpretation. For example, both
involved three or four different types of voice (bass, tenor, mezzo-soprano, and
soprano) and, as we have already seen, these may not use the same mechanism for
voice projection. Furthermore, while all of the singers in the second study had been
pupils of one teacher, this was apparently not true of the first. Consequently it is the
similarities rather than the differences between the studies that are likely to be most
significant. As the subjects were all classically trained professional singers, it is likely
that they were already using breathing strategies optimized for projection even in their
unsupported singing. This may make the critical mechanisms underlying each style
of singing harder to determine. From the viewpoint of young singers in training, an
examination of the evolution of respiratory patterns as technical facility is acquired
would be useful for guiding their development, but long-term studies of this nature
have not so far been carried out.

Wind Playing
Many of the most detailed published studies on respiratory patterns of wind players
also suffer from the very small numbers of subjects involved. This may in part be
related to the invasive nature of some of the measuring techniques used. Not many
players will relish performing with pressure balloons inserted via the nose down into
the stomach and esophagus. We cannot therefore be certain that the results so far
obtained will prove typical of the majority of elite performers. However, as more in-
formation is accumulated clearer trends should emerge, and even players of a profes-
sional standard may find it beneficial at least to experiment with different respiratory
strategies. Some studies have revealed considerable similarities between players at
very different levels of technical development in certain aspects of respiration, for
example, in the duration of breaths used for a given piece (Bouhuys 1964), though
these may not turn out to be the most significant parameters for performance. Other
studies have emphasized differences in the measurements obtained of the same pa-
rameter in a single player when practicing regularly compared to those obtained after
Figure 4-12
 Breathing in Singing and Wind Playing 127

a prolonged period without playing the instrument (Roos 1936). As with singers, it
is likely that differences in body structure will ultimately prove highly significant for
respiratory patterns.
The relationship between respiratory parameters and the type of sound produced
has not so far been examined objectively in wind players; for example, the studies
generally show no spectral analysis of the sound produced under different breathing
regimes or embouchure configurations. Clearly the type of sound that is deemed desir-
able or which is actually attained will be affected by many factors, such as the personal
taste and characteristics (e.g., embouchure configuration) of the individual performer,
the prevailing culture of players belonging to different schools of performance (here
the age of some studies may be a significant factor), and the characteristics of the
mouthpiece or the instrument. Some discrepancies between published studies have
been attributed to such factors (see Cossette, Sliwinski, and Macklem 2000). Because
a number of these reports are now quite old, differences in the lifestyle of the players
may also be significant. For example, the author of one of the few available large-scale
surveys of lung volume and breathing patterns in wind players reports that many of
the players were heavy smokers (Bouhuys 1964). That smoking has a dramatic effect
on lung function is now widely known and so it is unlikely to be as prevalent among
today’s players, who live in a more competitive and health-conscious society than did
those in the 1960s. The respiratory requirements demanded of wind players vary con-
siderably depending on the instrument. The most detailed studies performed to date of
the respiratory mechanisms of wind players have been carried out on flutists.

Respiratory Patterns in Flute Playing

Because the flute is played with an unstopped embouchure (i.e., with the mouth open),
its respiratory demands are more similar to those of singers than of players of reed and
brass wind instruments. In particular, it is only in the flute that respiratory braking is
likely to be a significant factor, as in other instruments there is significant resistance to
airflow at the mouthpiece or reed (fig. 4-12). Flutists use 70–85 percent of their total
vital capacity in playing and typically start a playing breath from 85 to 98 percent of
total capacity, though this will obviously depend on phrasing. On a theoretical basis,
we described three distinct patterns of expiration in the Konno-Mead diagrams of fig.
4-7: 1) the rib cage may be allowed to descend before the abdominal muscles are en-

Figure 4-12. Breath pressure and airflow in wind playing. This diagram illustrates the demands
made on the respiratory system by woodwind and brass instruments and has been produced from
numerical data published by Pawlowski and Zoltowski (1987). Each instrument is represented by
a different type of line. The black lines represent a note near the bottom of the register and the
gray lines, a note near the top. The beginning of each line represents the note played piano and
the end with the arrow, the note played forte. The horizontal dotted line marks the approximate
value for the elastic recoil pressure of the thorax at 100 percent vital capacity. The numbers in-
dicate the mean breath duration for each note (usually from four players). Note that some notes
in the upper register of the clarinet require less breath pressure than those in the lower register.
Among the woodwinds, the oboe requires high pressure but low flow rates, while the flute re-
quires only low pressure but high flow rates in the upper register. For brass instruments, both
pressure and flow rate are high in the upper registers.
128 Chapter Four

gaged; 2) abdominal contraction may precede the descent of the ribs; or 3) contraction
of the abdominal muscles and descent of the rib cage may occur simultaneously. One
study of three professional flutists demonstrated that each employed a different strat-
egy, with no obvious differences in facility or musical outcome (Cossette, Sliwinski,
and Macklem 2000). There is therefore no clear evidence at present that any one of
these approaches provides a clear advantage over the others, though it must be borne
in mind that the investigation of the respiratory dynamics in wind instrumentalists is
still at the stage of baseline data collection, and the issue of how changing respiratory
patterns might affect different aspects of performance has not yet been addressed.
Players who are the subjects of these studies are generally described as being of pro-
fessional status, but this does not mean that they all make a similar sound. We also
have no knowledge of their basic lung function parameters or somatotypes, both of
which are significant omissions.
During playing, the flutist must regulate both the flow rate (the volume exhaled
per second) and velocity of the air passing through the embouchure. When playing at
the bottom of the register, flow rate and air pressure at the embouchure are both low.
As a consequence, in the first stages of an expiration, when the lungs are full and the
elastic recoil forces of the thorax are high, considerable respiratory braking by the
paradoxical activity of inspiratory muscles (particularly the intercostals) is needed for
breath conservation. As the lungs empty and the recoil forces decline, the active expi-
ratory muscles of the abdomen become increasingly important to maintain sufficient
airflow to support the note. With rising pitch, air pressure at the embouchure must be
increased, typically by a factor of 3.5–5 over the three octaves. This is accompanied
by a forward projection of the lips to decrease the length of the air jet (Fletcher 1975).
The velocity of the airstream increases by a factor of about three over the same pitch
range; however, because the area of the embouchure opening is correspondingly
reduced (by a factor of 2–3), the flow rate (i.e., the amount of air used per second)
remains almost static between D4 and G6, though this rises sharply by C7. Increasing
sound volume is supported by rising flow rate and greater embouchure area, accompa-
nied by smaller changes in velocity. The velocity increase with volume declines with
rising pitch, particularly in the third octave, where high velocities are required even
for relatively low sound intensities.

Vibrato
An important element of the sound that must also be controlled by the respiratory
system is vibrato. This is a rapid fluctuation in air pressure leading to an oscillation
mainly in intensity, though also to some extent also in pitch. The intensity fluctuations
are more pronounced in the upper partials, producing a fluctuation in timbre (Fletcher
1975). Vibrato is particularly important for the flute, as it generates fewer overtones
than other woodwind instruments (Bonard 2001) and in the absence of vibrato can
sound harsh and lacking in color. In the following account it is important to bear in
mind that even a cursory comparison of performances by well-known soloists will re-
veal very different patterns of vibrato, a factor that is not generally taken into consid-
eration in studies of wind players or singers. Instructions to flutists on how to produce
vibrato often refer to the use of the diaphragm. In some players at least, short repeated
 Breathing in Singing and Wind Playing 129

contractions of the diaphragm do indeed contribute to the production of vibrato

through an oscillation in respiratory braking. In the above-mentioned study of three
flutists (Cossette, Sliwinski, and Macklem 2000), each was asked to hold a sustained
note with vibrato for a period of ten seconds. One of the flutists used the diaphragm
rhythmically for the first half of the breath, but its activity then rapidly declined (pre-
sumably as the thoracic recoil forces faded), to be replaced with a rapid oscillation of
the abdominal expiratory muscles superimposed on their tonic level of activity. The
other two flutists appeared to use neither the diaphragm nor the abdominal muscles
and it was inferred that the vibrato was produced by oscillatory movements of the
muscles of the larynx, which would cause rapid fluctuations in the width of the open-
ing between the vocal cords. This contradicts the belief of many flutists that the larynx
is not involved in the production of vibrato; however, direct observation of the glottis
(the aperture between the vocal cords) reveals that movements of the vocal folds are
probably an almost universal feature of flute vibrato technique (King, Ashby, and Nel-
son 1988). Even some of the players in this study had previously considered that the
diaphragm was the main source of their vibrato. This has been confirmed by several
other investigators. The confusion concerning the source of vibrato is perhaps not sur-
prising given the paucity of sensation from the diaphragm and the uncertainty about
the sensory supply to the muscles of the larynx. Other players and teachers are more
aware of the origin of vibrato, however, and their descriptions of how it should be
produced refer to the glottis, throat, or larynx as well as the muscles of the abdominal
wall. The frequency of vibrato is generally around 5 hertz; the constancy of this figure
between players has led to the suggestion that it represents a resonance phenomenon
related either to the mechanical properties of structures that produce it or their control
by the nervous system. Though there is as yet no direct evidence for this proposition,
the fact that the frequency of vibrato is typically very similar between singers provides
some support for it (Sundberg 1999).
One situation in which the action of the diaphragm is both unequivocal and con-
sistent between performers is in the playing of a series of untongued staccato notes
(Cossette, Sliwinski, and Macklem 2000). Here each note is initiated by abdominal
contraction and terminated by a burst of intense activity in the diaphragm, which
rapidly reduces air pressure. This is therefore similar to the situation seen in singers
performing the Renaissance-style trillo described previously, though the frequency of
the notes generated in the flute study was much slower. However, this is not a task
that is usually demanded of players.

Reed and Brass Instruments

Brass and other woodwind instruments have a relatively high pressure threshold for
initiation of the vibration of the reed or the lips (figs. 4-12, 4-13). Even at the bottom
of their range, the threshold pressure required is above that generated passively by the
elastic recoil of the respiratory system at total lung capacity (Bouhuys 1964; Rahn
et al. 1946), meaning that respiratory braking is not normally an issue. A possible
exception is found in large-bore brass instruments such as the tuba, which appear to
make exceptional demands on the respiratory system in terms of the range of respira-
tory pressures required and the percentage of vital capacity used (figs. 4-12, 4-13).
Figure 4-13. Wind instrument breath pressure requirements. This
diagram demonstrates the pulmonary pressure generated and the
percentage of vital capacity used by wind players when maintaining
a note for the longest time possible. The thin dotted line indicates
the average pressure generated relative to the percentage of vital
capacity and is derived from a study of respiration that used nonwind
players as subjects. The dashed line that lies to the right indicates the
performance achievable only by the top 5 percent in this study. Oboe
players can maintain a single breath for very long durations because
their instrument requires only low pressures and small airflow rates.
The main limitation to the length of the note is the buildup of carbon
dioxide in the lungs. The piccolo and tuba players use up their vital
capacity rapidly because the mouth is unstopped in the former case,
and the bore of the instrument is very large in the latter. The diagram
demonstrates that brass players must develop a level of respiratory
performance that is considerably above the average found in the
general population. In fact, the pressures needed for playing loudly in
the upper registers may approach the limit of what is physiologically
possible. Reproduced with permission from Bouhuys (1964). N.B.
The pitch and volume of the notes played were not indicated in the
original study.
 Breathing in Singing and Wind Playing 131

Table 4-1. Mean Intrapharyngeal Pressure for Generating Notes with Different Brass
Instruments
64 Hz 128 Hz 256 Hz 512 Hz 1024 Hz
(C2) (C3) (C4) (C5) (C6)
Trumpet n.a. n.a. 20 40 120
French horn 35 40 50 110 >150
Trombone 10 15 16 22 70
Tuba 6 8 15 22 n.a.

Pressure units = cm H20

n.a. = not applicable, that is, outside the range of the instrument
Reproduced with permission from Stasney, Beaver, and Rodriguez 2003.

The pressure needed to initiate a note increases with frequency for most of these in-
struments, though in the clarinet it actually decreases in the upper register (Fuks and
Sundberg 1999). While increasing the air pressure at the embouchure generally results
in a note at a given pitch becoming louder, this tends to follow an asymptotic curve so
that it ultimately reaches a limit above which large increases in pressure produce little
effect in volume (Fletcher and Tarnopolsky 1999). The highest pressures required
among orchestral wind players are found in the upper registers of the trumpet, French
horn, and oboe. Professional players can generate pressures considerably in excess of
diastolic blood pressure (the peak pressure generated by the heart during a beat). For
example, while diastolic pressure in a healthy individual is typically around 19 kilo-
pascals, trumpeters playing loudly at the top of their range can generate intrathoracic
pressures of up to 25 kilopascals.
It has been suggested by some brass teachers that the pressure within the airway
needed to generate a given pitch is approximately the same regardless of the instru-
ment used. Only one study has investigated this hypothesis directly (Stasney, Beaver,
and Rodriguez 2003), and its results are shown in table 4-1. Unfortunately, only one
player of each instrument was assessed in each case, but three were section leaders in
a major professional orchestra and one, a master’s-level student. The results clearly
point to considerable differences in the pressures needed to produce the same pitch
across the brass family, with the highest pressures being required by the French horn,
and successively lower pressures in trumpet, trombone, and tuba. Presumably this
relates to the size of the mouthpiece and the depth and tension of the region of the lip
that is vibrating, but the precise details await further investigation. It was also found
that playing with poor technique increased the pharyngeal pressure required for a
given note by about 20 percent. Though it was not made clear what instructions were
given to the players to simulate poor technique, direct visualization of the pharynx
using a laryngoscope revealed that this involved a retraction of the tongue and an
increase in tension in the muscles in and around the larynx. Clearly these important
issues merit further study.
Wind instruments also vary considerably in the amount of air expended during
playing (Bouhuys 1964). Among the woodwinds, the airflow rates are lowest for the
oboe and cor anglais, which allow passages to be sustained for prolonged periods
under a single breath (fig. 4-13) even without resorting to special techniques such as
circular breathing. The limitation to the duration of the breath in this situation is the
132 Chapter Four

buildup of carbon dioxide in the bloodstream, which increases the drive to inhale. Low
flow rates are not a feature of all double-reed instruments, as in bassoon players they
exceed levels seen in single-reed instruments. The flow rates required by brass instru-
ments when playing at high volume in the upper register are also very high, increasing
with the bore of the instrument, and so are greatest in the bass tuba.

Long-Term Effects of High Pressure on Lung Function

The question of whether the high lung pressures needed to play many wind instru-
ments have long-term consequences for lung function has long been a contentious
one, and the increasing ease with which it is possible to gather sophisticated informa-
tion on respirometry has done little to resolve the matter. This may indicate that there
are numerous factors related to lifestyle, body morphology, and clinical history that
dominate the results and obscure any effect of playing on the data. For example, the
increasingly sedentary lifestyles and the rising incidence of asthma in Western soci-
ety are likely to prove highly significant. A study of one group of young adult brass
players (mean age twenty years, with an average of nine years’ playing experience)
showed a 15 percent greater vital capacity than age-matched controls, mostly due to
a large increase in expiratory reserve volume (Tucker, Faulkner, and Horvath 1971).
Earlier work by Bouhuys (1964) also suggested an increase in vital capacity in wind
players. Recent studies have concentrated on more dynamic parameters of respira-
tion that are now used in the diagnosis of obstructive and restrictive airway disease.
Obstructive conditions include asthma, in which the airways of the lungs become
constricted, while restrictive diseases lead to a reduction in total lung capacity. In
individuals who are not suffering from some overt lung disease, obesity can be a sig-
nificant cause of restrictive lung function (Spiegel et al. 1988). The lung function in
a group of male nonsmoking Turkish naval band members of a similar age and level
of experience to those in the study by Tucker et al. was compared with a group of
naval nonmusicians (Deniz et al. 2006). This suggested that the bandsmen exhibited
a highly significant reduction in many respiratory parameters such as peak expiratory
flow rate (PEF) and forced expiratory flow rate (FEF) at various points during the
exhalation, and the forced expiratory volume in the first second (FEV1). An observed
reduction in forced vital capacity (FVC) appeared to be correlated with the number
of years of playing. There was no significant difference between the woodwind and
brass players. All of the results were expressed as a percentage of the values expected
when corrected for age, sex, and height, helping to eliminate a range of confounding
factors. The authors considered the possibility that high-pressure trauma to alveoli
and small bronchioles might provide an explanation for the reduction in respiratory
parameters. However, in the absence of corroborating evidence from clinical follow-
up, they instead suggested that asthma, perhaps aggravated by the inhalation of lung
irritants during the deep breathing required for wind playing, might provide a more
likely explanation. Though the results of this study are striking, they are in marked
contrast to those of a comparison of professional North American wind players, string
players, and singers (male and female; average ages forty-one to forty-seven years)
that used similar methods (Schorr-Lesnick et al. 1985). This study revealed no dif-
ference between the wind players and the other two groups. For most parameters, the
 Breathing in Singing and Wind Playing 133

percentage values were close to those predicted on the basis of age, height, and gen-
der. One occupational group that shares the necessity to generate high lung pressures
with wind players are glassblowers. Though now a relatively uncommon profession,
several studies of lung function involving this group have been carried out. In one
that managed to track down eighty-five such individuals, forced vital capacity and
FEV1 were both significantly higher in the glassblowers compared to controls (Munn,
Thomas, and DeMesquita 1990), though some other studies have found no difference.
The question of whether a vocational requirement to generate high respiratory pres-
sures has an effect on lung function therefore remains unresolved. Though certain
forms of emphysema are associated with high pressure in the lungs and occupations
requiring prolonged respiratory effort, there is so far no evidence that this condition
is unusually prevalent in the players of high-pressure wind instruments (Gilbert 1998;
Harman 1998). The effects of the high pressures generated by wind players on other
tissues will be considered in chapter 6.

Role of the Larynx in Playing Woodwind and Brass Instruments

In reed instruments, the contribution of the larynx to the generation of vibrato is much
less consistent than in the flute and appears to be more a matter of individual tech-
nique than of necessity (King, Ashby, and Nelson 1988). As with brass instruments
(Ridgeon 1986), the flow of air is controlled by the size of the gap between the vo-
cal folds (the glottis) together with the degree of arching of the tongue, which alters
the area of the opening at the back of the mouth. The partial closure of the glottis
during the production of the note can be clearly seen using endoscopy (Weikert and
Schlomicher-Thier 1999). When players are encouraged to “relax the larynx,” what
is actually being encouraged is probably a relaxation of the muscles of the pharynx,
which lies above it, as this can also be observed when the area of the glottis is reduced
during playing. The vocal folds may also be brought forcibly together to support
heavy tonguing and for loud notes. The movements of the vocal folds are greatest in
saxophone players. Classical saxophonists generally use the larynx only to regulate
airflow (King, Ashby, and Nelson 1988), and neither this nor the diaphragm makes a
consistent contribution to the production of vibrato (Weikert and Schlomicher-Thier
1999). However, the vocal folds are used extensively by jazz players to modify the
sounds produced. In the “growl” technique, their movements may be forceful and as
a result of this style of playing, damage to the surface of the folds can induce hoarse-
ness of the voice.
In some players of brass and woodwind instruments, the larynx may close inappro-
priately when high pressure is generated in the airway and interfere both with airflow
and with articulation (Steenstrup 2004). This can lead to audible noises in the throat as
the air is forced through the closed glottis. The tight closure of the vocal folds under
these circumstances is called the valsalva maneuver and is seen in weight lifters who,
when making a lift, fill their lungs and close off the airway while forcefully contract-
ing the abdominal muscles. This helps to make the trunk rigid to support the weight.
Indeed, we may find ourselves doing this when lifting a heavy object. It also frequently
occurs in the rather more mundane process of defecation because it is by this means
that we raise abdominal pressure to expel feces. This is like squeezing toothpaste
134 Chapter Four

from a tube. Unfortunately for some wind players, the valsalva maneuver can become
engaged when high pressure is required to support the playing of a note. This is, of
course, counterproductive because it reduces the airflow into the instrument, but it may
have become a conditioned reflex because the individual has learned to do this in other
situations where it is appropriate. In such circumstances, a conscious effort is required
to avoid closing the vocal folds during playing until this becomes second nature. Simi-
larly, the rise in airway pressure caused by blocking the airflow with the tongue during
articulation may also trigger the valsalva maneuver, which can delay the onset of the
attack. The player may not be aware of the true origin of the problem but attribute it to
poor control of the tongue.
The level at which the larynx is held is also characteristic for different woodwind
instruments. Though the tone quality is mainly a product of embouchure, breath sup-
port, and the quality of the reed, resonances in the pharynx and larynx may also con-
tribute and help to give the sound made by different players its individuality (King,
Ashby, and Nelson 1988). This is discussed further in chapter 6. The contribution of
the larynx to the sound is probably least in flute playing, in which it remains fixed in
a neutral position. In single-reed instruments the larynx tends to rise with the pitch
of the note, though the degree of movement depends considerably on the individual
player. In double-reed instruments the larynx is held low and this may favor the gen-
eration of characteristic overtones in these instruments.

Circular Breathing
One of the characteristics of wind playing is the necessity to break the flow of the
music into phrases in order to take a breath. While this sometimes is used as a form
of punctuation that contributes to musical expression, in many instances it is driven

Figure 4-14. Normal and circular breathing in wind instruments. A. During normal inhalation,
air is drawn in through both the mouth and nose. The muscles of the soft palate are relaxed to
allow free airflow by both routes. B. During playing, the soft palate is raised and held in contact
with the back of the throat. This forms a seal, allowing the pressure in the airway to rise suffi-
ciently to generate the sound. C. During inhalation for circular breathing, the soft palate is pulled
firmly down against the raised back of the tongue, sealing the mouth chamber. The small volume
of air in the mouth is then compressed by the muscles of the cheek to maintain the sound, while
air is drawn rapidly into the lungs through the nose.
 Breathing in Singing and Wind Playing 135

purely by necessity. One way around this restriction is to use the technique of circular
breathing, which allows a continuous flow of air to be maintained from the embou-
chure. It is regarded as a specialized technique by most classical and jazz musicians,
and even those who have developed this skill are likely to use it relatively sparingly.
In some instruments, however, most notably the didgeridoo, it is an integral part of
standard playing technique. The mechanism of circular breathing is shown in figure 4-
14. Prior to starting to play, air is normally taken in through both the nose and mouth.
The soft palate is allowed to hang free (fig. 4-14A), neither in contact with the back
of the throat nor the tongue, allowing the passage of air into the throat by both routes.
When playing, the soft palate is pulled up against the back of the throat, creating a
seal that allows the pressure in the airway to rise sufficiently to drive the vibration of
the lips or reed (fig. 4-14B). If players suffer loss of seal (see chapter 6), this is due to
weakness of the muscles of the soft palate, which hold it in this position. In circular
breathing, as the lungs empty during the breath, the cheeks are allowed to puff out
like a pair of bellows to form a small air reservoir. During the switch to the following
inhalation, the soft palate is pulled downward against the tongue to block off the back
of the mouth (fig. 4-14C). While air is taken in through the nose, the cheek muscles
(the buccinators) contract to compress the air in the mouth, raising the pressure suf-
ficiently to maintain the sound. As the mouth reservoir is small, the intake of breath
must be rapid. This feat of coordination takes considerable practice. One of the most
difficult aspects is maintaining a consistent pressure when switching between the lung
and the mouth phases of the breath so that the transition is not marked by a dip in
sound intensity or pitch. Circular breathing is best suited for instruments that require
high pressures and low airflow rates, such as the double-reed instruments, the French
horn, and the trumpet. The next easiest are the other brass instruments and the single-
reed woodwinds, though it becomes progressively more difficult as the bore becomes
larger. The greatest difficulty is found in the flute, in which the mouth is unstopped,
but even with this instrument circular breathing is possible.

Wind Playing and Asthma

There is considerable anecdotal evidence for the beneficial effects of playing wind
instruments on childhood asthma, both in the scientific literature (Marks 1974; Gil-
bert 1998; Lucia 1994) and in the personal accounts of individual musicians, some
of whom have subsequently become prominent professional players. Though it does
not cure the condition, playing appears to reduce the symptoms and improve its man-
agement. Asthma episodes can be triggered by stress and exertion, so it may seem
paradoxical that wind playing can have a positive effect; however, though playing
requires breath control, it does not increase the heart rate to the levels typical of even
moderate exercise (Tucker, Faulkner, and Horvath 1971). A common breathing ex-
ercise recommended for asthma sufferers is a forced expiration through pursed lips
that in many ways resembles the action required to play a wind instrument. While
individuals with asthma may have difficulty in finding the motivation to carry out
such exercises conscientiously, they are a natural consequence of wind playing, and
there is a strong inducement to improve breath control and the duration of expiration
for musical ends. Children who are initially unable to hold a sustained note for more
136 Chapter Four

than a few seconds may be able to extend this to forty seconds or more over a period
of months. Of course, in the event of a severe asthma attack it is important to wait until
it is completely over before playing again, but audibly wheezing players who are not
otherwise in distress may continue (Marks 1974). Though there is no psychological
element in the origin of asthma as a disease, stress and panic may contribute to the
onset and progress of an attack. For young wind players, increased confidence in be-
ing able to control their respiration and manage their symptoms may be a significant
factor in their improvement (Lucia 1994). There is also circumstantial evidence that,
over a period of years, increased ventilation of the lungs contributes to the correction
of barrel- or pigeon-chest deformities in young asthmatic wind players who are still at
an age where the skeleton of the chest is capable of remodeling.

REFERENCES

Bonard, J. M. 2001. The physicist’s guide to the orchestra. Eur J Phys 22:89–101.
Bouhuys, A. 1964. Lung volumes and breathing patterns in wind-instrument players. J Appl
Physiol 19:967–75.
Campbell, E. J. M. 1970. Accessory muscles. In The respiratory muscles: Mechanics and neural
control, ed. E. J. M. Campbell, E. Agostoni, and J. Newsom Davis. London: Lloyd-Luke.
Cossette, I., P. Sliwinski, and P. T. Macklem. 2000. Respiratory parameters during professional
flute playing. Respir Physiol 121 (1):33–44.
Deniz, O., S. Savci, E. Tozkoparan, D. I. Ince, M. Ucar, and F. Ciftci. 2006. Reduced pulmo-
nary function in wind instrument players. Arch Med Res 37 (4):506–10.
Fletcher, N. H. 1975. Acoustical correlates of flute performance technique. J Acoust Soc Am
57 (1):233–37.
Fletcher, N. H., and A. Tarnopolsky. 1999. Blowing pressure, power, and spectrum in trumpet
playing. J Acoust Soc Am 105 (2 Pt 1):874–81.
Fuks, L., and J. Sundberg. 1999. Blowing pressures in bassoon, clarinet, oboe and saxophone.
Acustica 85:267–77.
Gilbert, T. B. 1998. Breathing difficulties in wind instrument players. Md Med J 47 (1):23–27.
Griffin, B., P. Woo, R. Colton, J. Casper, and D. Brewer. 1995. Physiological characteristics of
the supported singing voice. A preliminary study. J Voice 9 (1):45–56.
Harman, S. E. 1998. The evolution of performing arts medicine. In Performing arts medicine,
ed. R. T. Sataloff, A. G. Brandfonbrener, and R. J. Lederman. San Diego: Singular Publish-
ing Group.
Hoit, J. D., and T. J. Hixon. 1986. Body type and speech breathing. J Speech Hear Res 29
(3):313–24.
King, A. I., J. Ashby, and C. Nelson. 1988. Laryngeal function in wind instrumentalists: The
woodwinds. J Voice 1:365–67.
Konno, K., and J. Mead. 1967. Measurement of the separate volume changes of rib cage and
abdomen during breathing. J Appl Physiol 22 (3):407–22.
Lagefoded, P. 1967. Three areas of experimental phonetics. London: Oxford University Press.
Leanderson, R., J. Sundberg, and C. von Euler. 1987. Role of diaphragmatic activity during
singing: A study of transdiaphragmatic pressures. J Appl Physiol 62 (1):259–70.
Lucia, R. 1994. Effects of playing a musical wind instrument in asthmatic teenagers. J Asthma
31 (5):375–85.
 Breathing in Singing and Wind Playing 137

Marks, M. B. 1974. The 1974 Bela Schick Memorial Lecture. Musical wind instruments in
rehabilitation of asthmatic children. Ann Allergy 33 (6):313–19.
Miller, R. 1997. National schools of singing: English, French, German, and Italian techniques
of singing revisited. Lanham, Md.: Scarecrow Press.
Munn, N. J., S. W. Thomas, and S. DeMesquita. 1990. Pulmonary function in commercial glass
blowers. Chest 98 (4):871–74.
Orozco-Levi, M., J. Gea, J. Monells, X. Aran, M. C. Aguar, and J. M. Broquetas. 1995.
Activity of latissimus dorsi muscle during inspiratory threshold loads. Eur Respir J 8
(3):441–45.
Pawlowski Z., and M. Zoltowski. 1987. A physiological evaluation of the efficiency of playing
the wind instruments—an aerodynamic study. Arch Acoustics (Poland) 12:291–299.
Pettersen, V., K. Bjorkoy, H. Torp, and R. H. Westgaard. 2005. Neck and shoulder muscle
activity and thorax movement in singing and speaking tasks with variation in vocal loudness
and pitch. J Voice 19 (4):623–34.
Pettersen, V., and R. H. Westgaard. 2002. Muscle activity in the classical singer’s shoulder and
neck region. Logoped Phoniatr Vocol 27 (4):169–78.
———. 2004. The association between upper trapezius activity and thorax movement in classi-
cal singing. J Voice 18 (4):500–12.
———. 2005. The activity patterns of neck muscles in professional classical singing. J Voice
19 (2):238–51.
Rahn, H., A. B. Otis, L. E. Chadwick, and W. O. Fenn. 1946. The pressure volume diagram of
the thorax and lung. Am J Physiol 146 (6):161–78.
Ridgeon, J. 1986. The physiology of brass playing. Manton, UK: Brass Wind Educational
Supplies.
Roos, J. 1936. The physiology of playing the flute. Arch Néerl Phon Expér 12:1–26.
Sand, S., and J. Sundberg. 2005. Reliability of the term “support” in singing. Logoped Phoniatr
Vocol 30 (2):51–54.
Schorr-Lesnick, B., A. S. Teirstein, L. K. Brown, and A. Miller. 1985. Pulmonary function in
singers and wind-instrument players. Chest 88 (2):201–5.
Spiegel, J. R., R. T. Sataloff, J. R. Cohn, and M. Hawkshaw. 1988. Respiratory function in
singers: Medical assessment, diagnoses, and treatments. J Voice 2 (1):40–50.
Stasney, C. R., M. E. Beaver, and M. Rodriguez. 2003. Hypopharyngeal pressure in brass musi-
cians. Med Prob Perform Artists 14:153–55.
Steenstrup, K. 2004. Teaching brass. Aarhus, Denmark: Royal Academy of Music.
Sundberg, J. 1999. The perception of singing. In The psychology of music, ed. D. Deutsch. San
Diego: Academic Press.
Thomasson, M., and J. Sundberg. 2001a. Consistency of inhalatory breathing patterns in pro-
fessional operatic singers. J Voice 15 (3):373–83.
———. 2001b. Lung volume levels in professional classical singing. Logoped Phoniatr Vocol
22:61–70.
Thorpe, C. W., S. J. Cala, J. Chapman, and P. J. Davis. 2001. Patterns of breath support in
projection of the singing voice. J Voice 15 (1):86–104.
Titze, I. R. 2000. Principles of voice production. 2nd printing. Iowa City, Iowa: National Cen-
ter for Voice and Speech.
Tucker, A., M. E. Faulkner, and S. M. Horvath. 1971. Electrocardiography and lung function
in brass instrument players. Arch Environ Health 23 (5):327–34.
Vilensky, J. A., M. Baltes, L. Weikel, J. D. Fortin, and L. J. Fourie. 2001. Serratus posterior
muscles: Anatomy, clinical relevance, and function. Clin Anat 14 (4):237–41.
138 Chapter Four

Waersted, M., R. A. Bjorklund, and R. H. Westgaard. 1994. The effect of motivation on shoulder-
muscle tension in attention-demanding tasks. Ergonomics 37 (2):363–76.
Watson, P. J., and T. J. Hixon. 1985. Respiratory kinematics in classical (opera) singers. J
Speech Hear Res 28 (1):104–22.
Weikert, M., and J. Schlomicher-Thier. 1999. Laryngeal movements in saxophone play-
ing: Video-endoscopic investigations with saxophone players. A pilot study. J Voice 13
(2):265–73.
Widmer, S., A. Conway, S. Cohen, and P. Davies. 1997. Hyperventilation: A correlate and
predictor of debilitating performance anxiety in musicians. Med Prob Perform Artists
12:97–106.
Wilder, C. N. 1983. Chest wall preparation for phonation in female speakers. In Vocal fold
physiology: Contemporary research and clinical issues, ed. D. M. Bless and J. H. Abbs. San
Diego: College Hill Press.
Winckel, F. 1952. Electroakustiche Untersuchungen an der Menschlichen Stimmen [Electro-
acoustic Investigations of the Human Voice]. Folia Phoniatr (Basel) 4:93–113.
 Chapter Five

The Voice
Management and Problems

The object of this chapter is to describe the origin of the voice. It will concentrate on
how the larynx produces sound and how this is modified by the resonant qualities of
the throat (the pharynx) and the mouth (oral cavity) during singing.

THE RELATIONSHIP BETWEEN

THE LARYNX, THROAT, AND MOUTH

The trachea and the esophagus pass down the neck together. The trachea is the route
by which air passes into the lungs; in order to keep the airway open, its walls are rein-
forced by horseshoe-shaped rings of cartilage. The esophagus lies immediately behind
the trachea. Waves of contraction run down its muscular walls to propel food into the
stomach. It is unsupported by cartilage, so when it is empty it is pressed flat between
the trachea, which lies in front, and the vertebral column and the muscles attached to it
(see chapter 2), which lie behind. The trachea and esophagus both open into the back
of the throat (also known as the pharynx), which is the common conduit for both air
and food (fig. 5-1). It is vital that food and drink not enter the trachea during breathing,
and the larynx evolved primarily as a valve to prevent this from happening. It sits at
the top of the trachea and in humans, its upper opening has a flap called the epiglottis,
which is pushed downward by the tongue during swallowing to block off the top of
the airway (animation 5-1).
The pharynx, like the esophagus, has muscular walls, but these do not collapse
because they are continuous with the cavities of the mouth and nose, which are sup-
ported by the bones of the skull. The wall of the pharynx has two muscular layers,
an inner one whose fibers run longitudinally (vertically) along its length and an outer
one whose fibers run circumferentially. Some of the longitudinal muscle fibers are
attached to the palate (the roof of the mouth) and others to the base of the skull or to
the opening of the eustachian tube, which runs from the back of the throat to the cav-
ity of the middle ear (see chapter 9). When the longitudinal muscle layer contracts,
the pharynx becomes shorter and wider. It is also drawn upward during swallowing

139
Figure 5-1. The larynx and pharynx. A. A view of the head and neck showing the different re-
gions of the pharynx and their relationship to the airway. The skeleton of the larynx is composed
of the cricoid and thyroid cartilages. The top of the airway is blocked by the epiglottis when it
is pushed backward by the tongue during swallowing. B. Enlarged view of the inner surface of
the larynx showing the vocal fold (vocal cord) and the vestibular fold with the slitlike cavity of
the ventricle lying between them. C. A view into the mouth showing the tonsils. The hard palate
forms the bony roof of the mouth. This continues as a muscular shelf known as the soft palate
from which hangs a central flap called the uvula.
 The Voice 141

by a muscle that runs from the base of the skull to the outside wall of the pharynx
(stylopharyngeus—fig. 5-2). The outer muscular coat of the pharynx is made up of
three sheets of muscle that sit one above the other. When they contract, these muscles
make the pharynx narrower and longer and hence are called the superior, middle,
and inferior constrictor muscles. The upper constrictor is attached at the front to the
muscular wall of the cheek, the middle constrictor to a bone at the root of the tongue
called the hyoid (see below), and the inferior constrictor to the cartilages of the larynx
(fig. 5-2).
The upper part of the pharynx can be divided into the oropharynx (mouth pharynx),
which encompasses the region below the soft palate that opens into the mouth, and the
nasopharynx (nose pharynx), which lies above the soft palate and opens into the back
of the nose. If you look into your mouth using a mirror, you can recognize the soft pal-
ate because of the small, fleshy flap (the uvula) that hangs down from the middle of it
(fig. 5-1). Farther forward in the mouth, the soft palate gives way to a horizontal shelf
of bone called the hard palate. The oropharynx is a passageway for food and drink as
well as for air, while the nasopharynx carries only air and is also the route by which
sound emerges in humming or during the production of certain consonants (e.g., m
and n). The soft palate is formed from two pairs of muscles. One pair runs horizontally
to meet in the middle, where the muscles are joined together. When these muscles
contract the soft palate becomes taut. They are therefore named the tensor palati. The
second pair of muscles (the levator palati) raise the soft palate to bring it into contact
with the back of the pharynx, sealing off the nasal cavity (animation 5-1, fig. 5-12). If
you use the mirror once more to look into your mouth while consciously closing the
back of the throat to block off the nose (as you might have done as a child when being
given a dose of foul-tasting medicine!), you will see the soft palate, together with the
uvula, rise sharply upward to achieve this. The soft palate is held in this position dur-
ing the playing of wind instruments as these require high oral air pressures to drive the
vibration of the reed or lips. Any loss of efficiency of this seal allows air to escape into
the nose and has a dramatic effect on the ability to play. It is also possible to close off
the back of the mouth from the pharynx by drawing the back of the tongue up against
the soft palate. This leaves the airway between the nose and the trachea open. Closing
off the mouth in this way to form a small air reservoir is used in circular breathing, a
technique used by some woodwind players to extend the duration of the breath (see
chapter 4). The passage to the nose and mouth can be closed simultaneously by rais-
ing the soft palate and pushing back the tongue. This is the initial phase of the act of
swallowing, which serves to drive the food into the lower part of the pharynx, where
it is then carried to the stomach by way of the esophagus (animation 5-1).
The pharynx and larynx are lined by a thin sheet of tissue known as an epithelium,
which is composed mainly of tall columnar cells. This is kept moist by watery secre-
tions of the mouth (from the salivary glands) and the nose. The epithelium also contains
specialized cells (known as goblet cells) that secrete mucus. For this reason it is some-
times called the mucosal epithelium or more simply, the mucosa. In histological terms,
it is described as a pseudostratified columnar epithelium with goblet cells. The mucous
film has the very important role of trapping dust particles in the airway to prevent them
from accumulating in the lungs. The film is moved around by the beating of minute
142 Chapter Five

Figure 5-2. Muscles supporting the larynx and pharynx and hyoid bone. These are active during
singing, regulating the level at which the larynx lies, its position relative to the hyoid bone, and
the shape of the pharynx. A. This diagram shows the central importance of the hyoid bone, which
lies at the root of the tongue. Many of the major muscles of the tongue (glossal muscles) and the
floor of the mouth are attached to it. It is suspended from the jaw and from a bony spine on the
base of the skull (the styloid process) that is also the origin of a muscle that raises the pharynx
(stylopharyngeus). The hyoid bone is attached to the larynx by the tissues of the pharynx and
by the thyrohyoid muscles and ligament. B. Muscles controlling the height of the larynx. These
are the thyrohyoid above and the sternothyroid below. A longer muscle, the sternohyoid, runs
directly from the sternum to the hyoid bone, while the omohyoid passes like a guy rope from the
hyoid bone to the top of the shoulder blade. The actions of some of these muscles are shown in
animation 5-1. Two nerves innervate the larynx. The superior laryngeal nerve supplies the crico-
thyroid muscle and carries sensation from the lining of the larynx above the vocal folds, while the
recurrent laryngeal nerve supplies all the other muscles and sensation below the vocal folds.

hairlike structures called cilia, which are found on the upper surface of the epithelial
cells. From the nose the mucus is propelled downward, while from the trachea and the
lungs it is propelled upward. Both streams ultimately reach the esophagus, where the
material is swallowed.

CONTROLLING THE HEIGHT OF THE LARYNX

If you were to look beneath the skin that covers the underside of the chin, you would
discover that the floor of the mouth is composed of a horizontal sheet of muscle (the
mylohyoid muscle) that runs from one side of the jaw to the other. Seen from the
inside, most of the floor of the mouth is taken up by the tongue, which plays an impor-
tant role in the articulation of words and in controlling the onset of the attack in wind
instruments. Several sets of muscles contribute to form the tongue. These are usually
 The Voice 143

divided into two groups, the intrinsic and the extrinsic muscles. The intrinsic muscles
are those that connect one part of the tongue to another, giving us great control over
its shape and the movement of the tip. The extrinsic muscles (fig. 5-2) suspend the
tongue between the palate (palatoglossus), the lower jaw (genioglossus), the base of
the skull (styloglossus), and the hyoid bone (hyoglossus). It is clear that the common
element of each of these names (glossus) refers to the tongue. These muscles allow the
tongue to be protruded or drawn back, raised or depressed. The organization of tongue
muscles may appear quite complicated at first, but one important point to note is the
central role of the small bone called the hyoid, which lies at the root of the tongue
and is shaped like a horseshoe with the opening facing backward (fig. 5-2A). It makes
no contact with the rest of the skeleton but is linked to many of the structures central
to the control of the singing voice. As well as anchoring many of the muscles of the
tongue, it is attached to the larynx through the relationship of the tongue to the muscu-
lar walls of the pharynx. The hyoid is also directly linked to the larynx by a membrane
that forms the wall of the airway at this point (the thyrohyoid membrane) and by a
pair of flat, straplike muscles, the thyrohyoid muscles (fig. 5-2B). A second pair of
straplike muscles attach the hyoid bone to the breastbone, or sternum (the sternohyoid
muscles), while a third pair (omohyoid) runs backward like a guy rope to the shoulder
blade. A final pair runs from the larynx to the sternum (the sternothyroid muscles).
These strap muscles control the absolute height of the larynx and its position relative
to the hyoid bone, and this has an effect on the sound produced during singing (see be-
low). These muscles are also involved in the raising and lowering of the larynx during
swallowing. It is this action that enables the backward-moving tongue to make contact
with the epiglottis and close off the airway (see animation 5-1). The movements of
swallowing must be closely coordinated with the breathing cycle. Just before swal-
lowing, a small breath is taken to draw air into the lungs. This helps to ensure there is
a positive pressure in the airway when the larynx is closed by the epiglottis so that any
food or fluid that does manage to get into the airway can be blown out. The position
of the larynx may also be influenced by activity in the inferior constrictor muscle of
the pharynx and by the stylopharyngeus. It is therefore very mobile, something that is
amply borne out by studies of how its position changes when singing in different parts
of the range (Hurme and Sonninen 1995).

THE VOCAL TRACT

The sounds that we make in speech and singing are not produced by the actions of
the larynx alone but involve the entire vocal tract (fig. 5-3). In functional terms this
consists of the following:

• The generator, which comprises the lungs and the muscles that drive air out of
them and control the rate of airflow through the larynx.
• The vibrator, which is the larynx itself. The vibrating elements of the larynx are the
vocal cords (more correctly called the vocal folds—the term we will use from now
on). The pitch of the sound is determined by the degree of tension in the vocal folds,
144 Chapter Five

Figure 5-3. The elements of the vocal tract are here compared to a trumpet. The lungs constitute
the generator (the source of the airstream), and the vocal folds represent the vibrator or voice
source (performing the same task as the lips of a trumpet player). This sets up a complex set of
vibrations (A) that are amplified by the resonator. For the voice this resonator is the pharynx and
mouth, the equivalent of the tube and bell of the trumpet. This has a set of resonant frequen-
cies called formants (B). In the trumpet these can be altered by changing the length of the tube
through the use of the valves, while for the voice, a limited degree of adjustment can be made
to both the width and length of the pharynx. The articulator of the voice is formed by the lips,
tongue, palate, and cheeks. There is no equivalent for the trumpet; however, placing the hand or a
mute across or into the bell and moving it in and out can be used to produce a voicelike modula-
tion of the sound. The sound that emerges is a product of the voice source spectrum modified by
the vocal tract resonances. This is shown in C, which comes from a recording of a voice.

which is in turn controlled by the muscles inside the larynx; the greater the tension,
the higher the pitch of the voice. The sound made is little more than a buzzing, simi-
lar to that produced by a vibrating reed or the mouthpiece of a brass instrument. This
is called the voice source. The pattern of vibration is complex and made up of a char-
acteristic pattern of harmonics overlying the fundamental frequency of the vibration.
This makes a significant contribution to the timbre and hence the individuality of the
voice between singers and in different vocal registers of the same singer.
• The resonator, which gives the sound its richness. This is formed by the pharynx
and the cavity of the mouth. In terms of the effect it has on the sound, it is equivalent
to the tube and bell of a wind instrument.
• The articulator, which shapes the sound to produce the vowels and consonants that
are combined to form words. The articulator is composed of the soft palate, tongue,
 The Voice 145

cheeks, teeth, and the opening of the mouth. If you think of the way in which the
sound of a trumpet can be manipulated by bringing the hand or a mute down over
the bell and then up again, it will be clear how even very simple manipulations of
the articulator can influence the quality of the sound. The articulator of the human
voice is much more complex than this and can be minutely adjusted by the muscles
of the palate, tongue, and face to produce the articulations that are characteristic of
different languages and of different individuals.

The properties of the generator are dealt with in chapter 4. In this chapter we will
examine in turn the nature of the vibrator, the resonator, and the articulator. Though
these will be considered primarily from the viewpoint of speech and singing, the ele-
ments of the vocal tract also play an important role in the playing of wind instruments.
This is discussed in chapter 6.

The Vibrator (Larynx)

The soft tissues of the larynx (which include the vocal folds) are supported by a skel-
eton formed not of bone, but of cartilage (figs. 5-4, 5-5). The main body of the larynx
is formed from two pieces of cartilage. One of these, the cricoid cartilage, is shaped
like a signet ring with the larger surface facing backward. Above the cricoid lies the
thyroid cartilage, a folded shieldlike structure that is open at the back (fig. 5-5F). The
thyroid cartilage looks a little like the prow of a boat except that along the midline
fold, the upper surface descends sharply into what is called the thyroid notch (fig. 5-
4A). The thyroid cartilage has a hingelike connection with the cricoid that allows it to
rotate downward like the visor of a helmet. The thyroid cartilage is more prominent in
men, in whom it is also known as the Adam’s apple. As already mentioned, when we
swallow, access to the trachea must be blocked to prevent food or saliva from entering
it. While the epiglottis is mainly responsible for this in mammals, it is a late adaptation
of the larynx, and the constriction in the airway created by the vocal folds was the
original structure for this purpose. The use of the folds as a means of producing sound
for communication evolved only later.
It is possible to look down into the larynx from above using an instrument called
a laryngoscope. When we do this we see the narrow opening through which the air
flows, bounded on each side by the vocal folds (video 5-1, fig. 5-13A). This opening
is known as the rima glottis. Among musicians this is often abbreviated to “glottis” in
descriptions of the mechanisms underlying singing and wind playing, though strictly
speaking the term applies to the vocal apparatus of the larynx as a whole. The glottis
controls the flow of air that is used either to drive the voice or the vibrations of the
reed or mouthpiece of an instrument. Immediately above the vocal folds is a horizon-
tal slitlike space called the ventricle of the larynx (fig. 5-4B), above which are two
smoothly rounded ridges, the vestibular folds. These are sometimes called the false
vocal cords but they make little contribution to the voice except in some extreme
forms of vocalization.
Much of the inner lining of the larynx is composed of a sheet of mucosal epithelium
similar to that of the pharynx and nose, though its secretions are augmented by pockets
146 Chapter Five

Figure 5-4
 The Voice 147

Figure 5-4. A. A view of the intact larynx seen from the front. B. The inside of the larynx. C. A
section through the side wall of the larynx at right angles to B, showing the internal structure of a
vocal fold. The epithelium that covers the inside of the throat and airway contains tall cells with
cilia that move fluid over the surface and goblet cells that produce mucus. Secretions are also
produced by glands that lie beneath the epithelium. At the edge of the vocal folds this epithelium
is modified to form a layer of tough, flattened cells that are capable of withstanding the impact
when the two folds strike each other during vocalization. The epithelium is separated from the
muscular core of the fold by a support tissue called the lamina propria, which has three layers.
The outer layer is elastic and mobile and together with the epithelium forms what is called the
cover of the fold. The inner two layers are composed of fibers that run along the length of the
fold, forming the vocal ligament at the fold’s free edge. The core of the vocal fold is the thyro-
arytenoid muscle. The muscle fibers nearest the edge are sometimes described separately as the
vocalis muscle, however, some authors use the terms vocalis muscle and thyroarytenoid muscle
interchangeably.

of glandular tissue within the larynx. In places where the surfaces of the two vocal
folds come into forceful contact with each other during phonation, the tall columnar
cells of the epithelium are replaced by several layers of tough, flattened cells that are
more resistant to abrasion. Histologically this is described as a stratified squamous
epithelium. Even with this added protection, the forces involved in vocalization
may cause physical damage to the surface of the folds, particularly in singers whose
technique is imperfect. Just prior to the beginning of phonation, the folds are brought
firmly together. How strongly this is done determines the degree of attack at the be-
ginning of the note. One situation in which the vocal folds may be damaged is when
the attack is habitually more forceful than is necessary. In addition, the more rapid
closure of the vocal folds that is typical of very loud singing also increases the force
of impact and risk of damage. The epithelium of the larynx sits on a sheet of support-
ing tissue called the lamina propria. The properties of the lamina propria of the vocal
folds are important for the production of sound. It is usually described as having three
layers (fig. 5-4C). The superficial layer is composed of a loosely organized feltwork
of randomly oriented stretchy fibers made of a protein called elastin. The intermedi-
ate layer is composed primarily of the same type of fibers but is oriented so that they
run parallel to the edge of the vocal fold. The deep layer mainly contains similarly
oriented collagen fibers that are thicker and much more resistant to stretching, and
beneath this lies the muscle that forms the core of the fold. The deep and intermedi-
ate layers constitute what is known as the vocal ligament, which is one to two mil-
limeters thick but becomes broader at its ends where it is under greatest mechanical
stress during phonation (Titze 2000). The mucosa and the superficial layer of the
lamina propria (together known as the cover) have some independence of movement
from the underlying tissue; when the two vocal folds strike one another, a ripple runs
across these two layers, starting from the free edge of each fold (video 5-1, see below).
Unfortunately, there is some ambiguity in how the layers that make up the folds are
named, making it difficult to compare accounts of the mechanics of vocal fold vibra-
tion in different vocal registers. To help you navigate these treacherous waters, three
commonly used schemes are summarized in table 5-1.
We will now consider the structural elements of the larynx and their movements
in the context of singing (figure 5-5). The narrow edges of the vocal folds are sup-
148 Chapter Five

Table 5-1. Naming Schemes for Components of the Vocal Folds

Three-Layer Five-Layer Two-Layer
Scheme Scheme Scheme

}
epithelium
mucosa { superficial layer cover
vocal ligament { middle layer
deep layer
muscle muscle } body

Note: Words in italics are components of the lamina propria.

Reproduced with permission from Titze 1994.

ported by the vocal ligaments, which are elastic and lie at the upper opening of a
flattened cone-shaped membrane (the conus elasticus) that forms a wall to seal the
spaces between the cricoid and thyroid cartilages. The vocal ligaments run from the
thyroid cartilage at the front to the small but very important arytenoid cartilages at
the back. About a fifth to a quarter of the edge of the vocal folds is composed of a
rigid spur (the vocal process) that projects forward from the arytenoid cartilage, so
only the region that lies in front of this is supported by the vocal ligament and is able
to vibrate during phonation. The space between the vocal folds (the glottis) is opened
and closed by muscles acting on the arytenoid cartilages,which therefore provide one
of the keys to understanding how the larynx works. Viewed from above, the arytenoid
cartilages appear L-shaped, with the two Ls lying back to back (fig. 5-5F, video 5-1).
The forward running leg of the L (the vocal process) is attached to the vocal ligaments
while the other short leg (called the muscular process) sticks out to the side and is
attached to several of the laryngeal muscles. At the junction of the two legs of the L
is a vertical projection attached to muscles that prevents the two arytenoid cartilages
from being drawn too far apart. The arytenoid cartilages sit on the top of the cricoid
cartilage; for the sake of simplicity we will consider that they rotate mainly around a
vertical axis. The larynx contains many muscles (fig. 5-5E), but only one pair opens
(abducts) the vocal folds. These are called the posterior cricoarytenoid muscles, and
the name tells you that the muscles are attached at one end to the back (posterior) of
the cricoid cartilage and at the other end to one of the arytenoid cartilages. By pulling
on the back of the muscular process of the arytenoid cartilages, these muscles make
them rotate in a direction that opens the gap between the vocal folds (see animation
5-2). They are active when we inhale, and if both are paralyzed (a life-threatening
situation), air cannot reach the lungs. Two muscles attached to the anterior (front)
edge of the muscular processes of the arytenoid cartilages have the opposite action,
causing the vocal folds to come together (adduct) to close the glottis. These are the
lateral cricoarytenoid muscles (fig. 5-5A, animation 5-2). In this they are aided by the
transverse or interarytenoid muscle, which links the two arytenoid cartilages and pulls
them together (animation 5-3). Most of the mass of each vocal fold is composed of
a muscle running parallel to and in contact with the vocal ligament (fig. 5-5C). This
is the thyroarytenoid muscle. As its name suggests, it is attached at one end to the
thyroid cartilage, close to the midline, and at the other end, mainly to the muscular
processes of the arytenoid cartilages. One part of it forms a thin horizontal shelf in the
Figure 5-5. A summary of the structure of the larynx and its movements. Each element of the
figure shows two views of the larynx, one from above and one either from the side or from be-
hind. The cartilaginous skeleton of the larynx is labeled in F. It is composed of the thyroid carti-
lage, which looks like the prow of a boat, and the cricoid cartilage, shaped like a signet ring. The
two arytenoid cartilages move the vocal ligaments, which form the free edges of the vocal folds,
to open and close the glottis. The edges of the vocal folds are formed from both the arytenoid
cartilage and the vocal ligament, but only the part supported by the ligament vibrates during
phonation. A and B show the actions of two muscles (the lateral cricoarytenoid and transverse
arytenoid [interarytenoid] muscles) that bring the vocal ligaments together (adduction). C and D
show muscles that tense (thyroarytenoid/vocalis) or stretch (cricothyroid) the vocal folds and E
shows the muscles (posterior cricoarytenoids) that open (abduct) the vocal folds. The actions of
some of these muscles are shown in animations 5-2, 5-3, and 5-4.
150 Chapter Five

core of the fold, running along side of the vocal ligament; this part is often described
as a separate muscle, the vocalis (fig. 5-4C). Unfortunately, there is a tendency for au-
thors to use the terms thyroarytenoid and vocalis interchangeably, so it is sometimes
difficult to know precisely what is being referred to.
Unlike the rest of the thyroarytenoid muscle, the vocalis component is attached to
the vocal process of the arytenoid cartilage. There has been some dispute over whether
the vocalis can truly be regarded as being functionally distinct from the thyroaryte-
noid, but studies of distribution of fast and slow contracting muscle fibers appear to
support the idea (Sanders, Rai, et al. 1998). The vocalis muscle has been shown to
contain a very unusual type of slow-acting muscle fiber that is not seen in other la-
ryngeal muscles (Han et al. 1999). This type of fiber is said to be tonic, which means
that it has adapted to remaining in the contracted state for prolonged periods. It has
further been suggested that the vocalis component itself has two functionally distinct
parts or compartments. The lower compartment is said to be composed of a single
large muscle bundle, while the upper part is made up of many small, loosely packed
muscle bundles. Standard descriptions of the action of the thyroarytenoid muscle state
that it pulls the thyroid cartilage toward the cricoid, slackening the vocal ligament and
the cover, while the vocalis is said to slacken the posterior part of the fold and tighten
the anterior part. Under these circumstances, the vibration would be confined to the
section of the fold overlying the contracted region of the vocalis. This is analogous to
shortening the vibrating segment of a string, and therefore increases the frequency of
the sound generated. Some mechanism of this kind is needed to explain why the pitch
of the voice continues to rise even when the vocal folds can be stretched no farther.
This stretching is brought about by the cricothyroid muscle, which opposes the actions
of the thyroarytenoid muscle. The cricothyroid muscle runs between the cricoid and
thyroid ligaments at the side of the larynx. This rotates the thyroid cartilage toward the
cricoid cartilage, stretching the vocal folds (fig. 5-5D, animation 5-4). There is also
some disagreement as to whether this is achieved by an upward movement of the cri-
coid arch or a downward movement of the thyroid cartilage. Though some authors em-
phasize the movement of the cricoid, this is to some degree restricted by its attachment
to the first ring of tracheal cartilage. Furthermore, during phonation, the lamina of the
cricoid cartilage (the large, flat, backward-facing surface of the “signet ring”) is said
to be pressed firmly against the vertebral column, limiting its ability to rotate upward
(Gray and Williams 1989). On the other hand, though the laryngeal cartilage should
be more free to move, it is attached to the strap muscles at the front, and depending
on their state of contraction, these may aid or inhibit its forward and downward move-
ment. It therefore seems most likely that some degree of movement of both the thyroid
and cricoid cartilages may occur through the action of the cricothyroid muscle, but
that which moves most may depend on the state of other muscles.
The length of the vibrating part of the vocal folds (i.e., the part occupied by the
vocal ligaments) is difficult to measure in a living person without using X-rays or
other specialized equipment; however, it has recently been discovered that this has a
relatively fixed relationship to the distance between the lower border of the cricoid
cartilage and the thyroid notch (Williams and Eccles 1990). This external feature of
the larynx can easily be measured with simple instruments and can predict the funda-
 The Voice 151

mental frequency of the voice. If we call this measure M, the frequency is predicted
by the following equations:

Fundamental frequency (hertz) = 190 or (M  1.7) for men

Fundamental frequency (hertz) = 260 or (M  1.4) for women

The typical average fundamental frequency as revealed by a reading task is about 112
hertz (Hz) for males and 206 Hz for females. Though there is a significant overlap in
laryngeal size between males and females, studies involving large numbers of subjects
have revealed little or no overlap in fundamental frequencies between the sexes. Con-
sequently, other factors must contribute to the differences in fundamental frequency
between the sexes. The most likely explanation is that in the male, the vocal folds have
a greater mass per unit length (Williams and Eccles 1990).
The muscles of the larynx are supplied by two pairs of nerves (one pair on each
side). The cricothyroid muscle is controlled by the superior laryngeal nerve, which
runs down the side of the larynx, while all the other muscles are controlled by the
recurrent laryngeal nerve, which enters the larynx from below (fig. 5-2B). Unlike
most of the other muscles in the body, we do not have any conscious sensation of
the activity of the laryngeal muscles, so we are not aware when they are contracting.
Any sensation of tightness that is felt in the throat probably comes from the pharynx.
The lining of the larynx is very sensitive to touch, however; something brought to
our attention only too clearly when a fragment of food “goes down the wrong way.”
The region of the mucosa above the vocal folds is supplied by the superior laryngeal
nerve and the region below by the recurrent laryngeal nerve. The nerves supplying the
larynx are potentially vulnerable during surgery to the front of the neck (e.g., involv-
ing the thyroid gland). Damage to one of the recurrent laryngeal nerves can result in
paralysis of almost all of the muscles on that side of the larynx. The inability to move
the vocal folds results in a husky voice due to paralysis of the posterior cricoarytenoid
muscle, which cannot therefore open the fold on the affected side. The nerve branches
may be able to regain contact with the muscles as long as they can find a way back to
their appropriate targets, but this takes several weeks, as they grow at a rate of around
one millimeter per day, at best.

Movements of Vocal Folds

The classical description of the movement of the vocal folds during singing is as
follows. At the beginning of phonation, just before the air is driven from the lungs
through the larynx, the vocal folds are brought together (i.e., the glottis is closed). As
the air is then driven from the lungs against the closed folds, the subglottic pressure
(the pressure in the region below the glottis) rises until it is capable of pushing the
vocal folds apart. The edges of the vocal folds are elastic; as the airstream begins to
flow between them, they become progressively more stretched. The farther apart they
become, the greater are the elastic forces trying to close them again. Finally, this ex-
ceeds the subglottic pressure, and the folds snap shut and the cycle repeats itself (fig.
5-6, animations 5-5 and 5-6, video 5-1). Other factors also help to close the glottis.
Figure 5-6. The vocal folds. A and B show the movement of the vocal folds through one com-
plete vibratory cycle. The column in A shows a series of views of the vocal folds as seen from
above. B shows a series of transverse sections through the folds corresponding to each stage in the
cycle. In B the black patch indicates a ripple on the surface of the vocal fold that moves across
it during the cycle. In C, the black dot indicates the position on a flow glottogram (see fig. 5-8)
represented by each image in A. The movements of the folds can also be seen in animations 5-5
and 5-6 and video 5-1.
 The Voice 153

The faster the air rushes through the gap between the folds, the lower the pressure be-
tween them, a physical phenomenon known as the Bernoulli effect. This sucks the vo-
cal folds together and aids the elastic forces that are trying to close the glottal opening.
This description of the mechanism underlying the oscillation of the vocal folds does
not offer a complete explanation of the events that take place. In particular, it would
not explain how vocal fold oscillation can be maintained in a situation in which the
glottis does not close completely, for example, during breathy phonation (see anima-
tion 5-3). An important element of the mechanism that ensures the sustainability of the
oscillation is a process known as inertive reactance (Titze 2008). When the subglot-
tal pressure first drives the folds to open, the air passing between them encounters a
mass of stationary air in the space immediately above. It takes some time for this air
mass to accelerate to a steady velocity, and during this period, the barrier it presents
causes the pressure of the air being driven from below to rise. This helps to force the
folds open farther. Later, as the elasticity of the vocal folds acts to close again, the
inertia of this air mass means that it takes some time to decelerate. This leads to the
formation of a region of low pressure immediately above the folds that helps to suck
them closed again. Inertive reactance serves to give the oscillation a periodic kick,
analogous to pushing a swing in phase with its movement, which helps maintain the
oscillation. Readers with a mathematical bent will find a full description of this and
other processes that support vocal fold oscillations in Titze (2000).
The frequency of the opening and closing of the glottis determines the pitch of the
note, so if this happens 440 times a second (i.e., at 440 Hz), the pitch of the funda-
mental note produced will be A4. When the vocal folds come together to produce a
note of low frequency, they first make contact deep in the glottal opening. As a result
of the impact, a long ripple is generated on the upper surface of the vocal folds that
runs away from the opening, parallel to the free edge (fig. 5-6B, animations 5-5 and
5-6, video 5-1). This is called the mucosal wave and represents a displacement of the
cover (see table 5-1), which indicates that it is not under significant tension and is
only loosely attached to the underlying vocal ligament and muscle. The ripple plays a
significant role in making the vocal fold oscillation self-sustaining. If the wave is ab-
sent for some reason, considerably greater subglottal pressure is required to maintain
vocal fold vibration, particularly in the upper registers (Titze 2000). Current wisdom
is that the folds must be adequately hydrated to maintain the mobility of the cover and
consequently that the singer must take in a sufficient quantity of fluid on a daily basis
to ensure that this is the case.
At high frequencies (e.g., in the falsetto register), the tension of the vocal folds
is carried mainly by the vocal ligament and cover, with the result that no ripple is
generated. It has been proposed by some that the mobile cover part of the vocal folds
may contain the upper compartment of the vocalis muscle and that this is controlled
separately from the lower compartment (Sanders, Rai, et al. 1998). If this is true, it
implies that the upper compartment plays an active role in controlling the shape and
tension of the edge of the vocal fold, which in turn contributes to the control of pitch,
particularly in the falsetto register (see animation 5-4).
One way of displaying the airflow through the glottis during vocal fold vibration
is what is known as a flow glottogram (figs. 5-6C, 5-7, animations 5-5, 5-6). The
154 Chapter Five

velocity of the airflow fluctuates in time with each cycle of opening and closure of the
vocal folds. The louder we sing, the greater the peak flow rate during the open phase
and the more abruptly the flow is cut off as the vocal folds snap shut. There is a linear
relationship between the intensity of sound produced and flow rate—a doubling of
flow rate produces about a 9 decibel (dB) increase in volume (Bouhuys, Proctor, and
Mead 1966). However, as the flow rate increases, the intensity of the upper harmonics
rises to a greater extent than that of the lower harmonics. For example, an increase that
produces a rise in intensity of 10 dB at a fundamental frequency of 600 Hz, sung by
a baritone, generates a 16 dB increase in harmonics close to the singer’s formant at 3
kilohertz (kHz) (Sjolander and Sundberg 2004), which alters the timbre of the sound.
Increasing subglottal pressure and flow rate also causes a rise in the pitch of the note,
so to stay in tune during changes in volume, the singer must adjust the strength of
contraction in the muscles that close and tense the vocal folds. Because the degree of
opening of the folds is also related to their length, singers with longer vocal folds (i.e.,
those with low voices) have stronger fundamentals in the notes they produce than do
singers with short vocal folds.
The quality of the sound generated by air passing through the glottis varies de-
pending on the precise action of the vocal folds. If the vocal folds are held strongly

Figure 5-7. Flow glottograms are a measure of airflow through the glottis (the space between
the vocal folds). As the folds open and close at the fundamental frequency of the note sung, flow
glottograms show the rapid fluctuations in airflow. These idealized glottograms show that as the
singing gets louder, the folds closer ever faster and the open phase becomes shorter.
 The Voice 155

together, this results in what is known as “pressed phonation” (see animation 5-3). In
this mode, the fundamental frequency of the note generated is weak, giving the voice
a harsh sound due to the greater prominence of the upper harmonics. This is a result of
increasing the stiffness of a vibrating surface, which leaves it more prone to damping
the lower frequencies and better able to sustain higher frequencies. When the vocal
folds are not so strongly pressed together, the result is “flow phonation,” in which
the fundamental frequency is strong (up to 15 dB greater in intensity than in pressed
phonation). However, if the vocal folds are so weakly brought together that the open-
ing is not closed completely (animation 5-3), the fundamental is again weak and the
voice is described as “breathy.” Many young singers exhibit a degree of breathiness
in their phonation. This is normal and may be particularly associated with weakness
in the interarytenoid muscle, with the result that a triangular gap or “chink” is left
between the arytenoid cartilages during phonation. As the laryngeal muscles mature
and strengthen, the breathiness generally disappears. Most Western classical singers
base their singing voice on flow phonation, though the other types of phonation may
also be used for expression and in particular contexts. In speech, flow phonation is
characteristic of low pitch and quiet vocalizations, and pressed phonation with higher
and louder vocalizations such as shouting.
In whispering, the vocal folds do not vibrate and are kept partly open because the
posterior cricoarytenoid may be partly active and the interarytenoid muscle relaxed
(animation 5-3). The edges of the folds are either held straight for their entire length
like an inverted V or are first straight then angled outward toward the ends, which
attach to the arytenoid cartilages like an inverted Y whose stem is either closed
completely, or itself a narrow V (Solomon et al. 1989). As a result, most of the air
bypasses the vocal folds, and the sound is generated as a result of turbulence in the
airflow. The false vocal folds are often displaced downward in whispering and may
even be brought into contact with the true vocal folds, limiting their ability to vibrate
(Tsunoda et al. 1997).

Vocal Registers
Any account of how the larynx is used in singing must explain not only how the pitch
of a note is regulated, but also why the quality of the sound changes in different vocal
registers. A vocal register is a pitch range over which the timbre (harmonic richness)
of the sound remains relatively constant. The vocal range of a singer generally has
two or more registers; for example, in a male voice there is typically a modal register
(also known as the chest or normal register) and a falsetto register. In some musical
cultures (e.g., in Russia and Eastern Europe) a very low Strohbass (or pulse) register
is also used by the lowest male voices. For the female voice, a chest and a head reg-
ister are generally distinguished. The situation is made confusing by the presence of
several naming schemes. Table 5-2 summarizes the features of the main registers and
lists some of their more common synonyms.
The timbre depends on what elements of the vocal folds are vibrating; this is shown
schematically in animation 5-7. When most of the mass of the fold is in motion (the
muscle plus the overlying tissue), the timbre is rich in strong lower harmonics. This
is typical of the modal register. If the muscular core of the fold is allowed to relax so
Table 5-2. Main Register Features
Register Muscles Active Vibrating Region of Fold Voice Quality Pitch
Pulse (Strohbass) TA only Most of the thickness of the Pulsating; vocal fry Lowest; below normal singing
fold
Chest (modal) Mostly TA, some CT Most of the fold, both cover Heavier, richer tone; vibration Lower part of singing range
and body felt behind breastbone
Head (middle or mixed) Mostly CT, some TA Cover only tone; vibration Lighter, thinner, felt in head Upper part of singing range
Falsetto CT only Only outer cover layers Lightest possible Highest sung pitches; above normal
range

TA = thyroarytenoid; CT = cricothyroid
Reproduced with permission from Titze 1994.
 The Voice 157

that most of the tension is carried by the vocal ligament, then the harmonic structure
of the sound is depleted. This is typical of the falsetto register, in which the vocal
folds are also more stretched, less flexible, and thinner than in the modal register and
the glottis is rarely completely closed. The characteristics of the pulse register have
a different origin. When the folds are vibrating very slowly, the resonances set up by
each opening and closing event may die out before the next one occurs. The result
is that the sound is heard as a series of separate pulses said to be like the crackling
of breaking straw when it is walked upon (hence Strohbass) or like the sound of fry-
ing fat (vocal fry). This is generally heard only at pitches below 70 Hz (about C2)
and can take a number of forms (Scherer 2006). At very low pitch (e.g., around 30
Hz), the vocal fry may represent each single cycle of vocal fold opening and closing.
A roughness at higher frequencies may be the result of alternate cycles of different
amplitudes, producing an audible pulsation at half the frequency of the note; this is
the most typical form. More complex patterns can be produced in which the flow rate
through the glottis waxes and wanes at regular intervals through succeeding cycles.
An example shown by Scherer shows a low-amplitude pulse at 93 Hz superimposed
on a larger one at 7.5 Hz.
When crossing from one register to another, the voice of an inexperienced singer
may show a break, that is, an abrupt change in timbre, something that vocal training
aims to eliminate. One of the complicating aspects of registers is that because their
ranges overlap, a singer will often be able to produce a note of a particular pitch in two
or even three registers (Hirano, Vennard, and Ohala 1970; Hollien and Moore 1960).
One can liken this to playing the same note on two guitar strings, one of which is a
thin nylon monofilament string and the other a string overwound with metal wire to
increase its mass. The heavier string can match the pitch of a note played on the thin-
ner one by being shorter or under greater tension; however, the quality of the note is
not the same because they have different harmonic structures.

Laryngeal Muscle Activity during Singing

The precise way in which the muscles act during singing is controversial. Observing
the activity of muscles in the larynx, which are small and delicate, is difficult and in-
volves the placing of fine-wire hook electrodes into them. This is both uncomfortable
and potentially risky for singers and as a consequence there have been few investiga-
tions of this nature. Unfortunately the results of those that have been carried out are
not in complete agreement (Shipp, Doherty, and Morrissey 1979; Hirano, Vennard,
and Ohala 1970; Hillel 2001). For this reason we will start by considering the prin-
ciples underlying the control of pitch and register and then consider the experimental
evidence to support them.
During phonation, the vocal folds are brought together (adducted) by the lateral
cricoarytenoid muscles, which rotate the arytenoid cartilages toward each other. At
the same time, the interarytenoid muscle draws the arytenoid cartilages closer together
(animation 5-3) and the thyroarytenoid muscle contracts to bring the vocal folds under
tension. Electrical recordings show that these three sets of muscles are activated si-
multaneously immediately prior to vocalization, but that once this has started, activity
in the interarytenoid muscle dominates (Hillel 2001). In singing, the cricothyroid and
Figure 5-8
 The Voice 159

Figure 5-8. A. The relationship between activity in the thyroarytenoid and cricothyroid muscles
and the vocal pitch. This graph demonstrates how pitch can be increased by activity in either
the thyroarytenoid or cricothyroid muscles alone or by the two acting in concert. The degree
of muscular effort required to produce the note also depends on the air pressure generated by
the respiratory system. The lines represent the generation of sound at 200, 300, 400, and 500
Hz. Increasing subglottal pressure raises the frequency of the sound, so at higher pressures, less
muscular effort (less tension in the folds) is needed to generate a note of a given pitch. The qual-
ity (timbre) of the sound will depend on how the note is achieved. If the tension in the folds is
increased mainly by the thyroarytenoid muscle, a point is ultimately reached when the muscle
can contract no more. To raise the pitch further, the thyroarytenoid must be relaxed and the
cricothyroid brought into play (black arrows). This will lead to an abrupt change in timbre (i.e., a
register break) between the modal to falsetto registers. The trained singer (gray arrow) learns to
avoid this by continuously adjusting the activity of both muscles. The result is a gradual change
in timbre. B, C. Cross-sections through the vocal fold when singing in the modal or falsetto
registers. In the modal register, the thyroarytenoid muscle is contracted, which makes the edge
of the fold blunter in cross-section and increases the contact area between the two folds. The
entire thickness of the vocal fold vibrates and the sound is rich in low harmonics. In the falsetto
register, the thyroarytenoid relaxes and the cricothyroid contracts, the edge of the fold becomes
thinner as the tension is transferred to the vocal ligament, and the vibration is mainly in this and
the mucosa, leaving the sound with weak lower harmonics and hence brighter. Reproduced with
permission from Titze (2000).

thyroarytenoid muscles act to some extent antagonistically. The cricothyroid rotates

the thyroid cartilage downward and in so doing stretches the vocal folds. However, as
we have seen, these are made up of several components, including the thyroarytenoid
and vocalis muscles, the vocal ligament, and the mucosa. If the thyroarytenoid muscle
is relaxed when the folds are stretched, their edges become thinner and most of the
tension is carried by the thin vocal ligament (fig. 5-8A, B, animation 5-7). This can
vibrate at high frequency, and the timbre is light and typical of the falsetto register.
The thyroarytenoid muscle shortens the vocal folds and resists the action of the crico-
thyroid. When both muscles are active, most of the tension and vibration in the vocal
folds is transferred to the much more massive thyroarytenoid muscle (animation 5-7).
This favors lower fundamental frequencies and the rich timbre typical of the modal or
chest register (table 5-2, fig. 5-8A). The contraction of the thyroarytenoid also thickens
the edge of the vocal folds, increasing the contact area when the glottis is closed (fig.
5-8C). Figure 5-8 illustrates a theoretical explanation for the abrupt register transi-
tion seen in untrained singers (Titze 2000). Both the cricothyroid and thyroarytenoid
muscles are active in the modal and falsetto registers, and to achieve a particular timbre
what is important is not the absolute level of activity in each muscle but the relative
activity in the two muscles. Thus the same pitch can be achieved in a variety of ways,
but the timbre will depend on which muscle is dominant. It is hypothesized that in the
untrained singer, activity in the thyroarytenoid dominates in the modal register. This
produces a rich timbre but when the muscle is fully active, the pitch can increase no
further in this register. To raise the pitch, the thyroarytenoid must be relaxed abruptly
to allow the cricothyroid to increase the tension in the vocal fold further. Over a very
small frequency interval, the tension in the vocal fold is switched from the muscle to
the vocal ligament, leading to a noticeable change in timbre characteristic of a register
break (black arrows). It is envisaged that with training, the singer learns to modulate
160 Chapter Five

the pitch by a more gradual change in the pattern of activity in the two muscles, which
leads to a smooth and continuous change in timbre throughout the vocal range (gray
arrow).
Experiments that directly monitor the activity of laryngeal muscles through im-
planted electrodes (Hillel 2001; Hirano, Vennard, and Ohala 1970) confirm that
cricothyroid activity increases as pitch rises within a given register. Noninvasive ul-
trasound measurements of changes in the dimensions of the space between the cricoid
and thyroid cartilages are consistent with this observation (Laukkanen et al. 2002).
This confirmation is important, as the risk that direct recordings from the muscles may
alter the way they are used is a significant one. However, the downside of ultrasound
studies is that the evidence of muscle activity is rather indirect, so it is important that
the two approaches be used in concert. Hirano concluded that activity in the lateral
cricoarytenoid (which closes the gap between the vocal folds) also increases with
pitch and that this appears to be matched by activity in the interarytenoid muscle,
which draws the two arytenoid cartilages together. Both therefore act to keep the
glottis closed and will also help to prevent the arytenoid cartilages from being pulled
forward by the action of the cricothyroid and thyroarytenoid muscles. We previously
simplified our consideration of the movement of the arytenoid cartilages by describing
only rotation around a vertical axis, but the shape of the surfaces of their joints with
the cricoid cartilage also allows them to slide forward.
Hirano et al. reported that when the limit of the register is reached and the voice
switches to a higher register, activity of both the cricothyroid and the vocalis muscle
declines sharply. It is probable however, that what they meant by the term vocalis was
the thyroarytenoid muscle, as it is almost inconceivable that the recording techniques
used could access the vocalis selectively, and no mention is otherwise made of the thy-
roarytenoid. Relaxation of the thyroarytenoid would reduce the resistance of the vocal
folds to stretching under the action of the cricothyroid muscle and allow the vocal fold
to become thinner. As a result, tensioning the fold to a given pitch would require less
effort by the cricothyroid and might explain the reduction in its activity. Though this
is not predicted by figure 5-8A, which is intended as a simple diagram to illustrate a
theory of register change, it is not inconsistent with the ideas on which it is based.
The ultrasound studies of Laukkanen (2002) found that the pitch can still rise after
the vocal folds have reached their maximum level of stretch. This must therefore be
achieved by other mechanisms. One would be to increase the airflow through the
glottis, which, as we have already seen, increases pitch. Another is the one previously
mentioned, that in the upper falsetto register, the posterior parts of the vocal folds are
held together and only the anterior parts are allowed to vibrate, that is, as the vibrating
surfaces effectively become shorter the frequency would rise. For this to be possible,
different parts of the vocalis muscle would have to be under separate control (Sanders,
Rai, et al. 1998), for which there is as yet no direct evidence.
Though the description of the activity patterns in the laryngeal muscles by Hirano
et al. appears quite rational, a second study of laryngeal muscle activity failed to
show similar patterns during register changes (Shipp, Doherty, and Morrissey 1979).
This might indicate that there are several ways of achieving the same vocal goals
(Sundberg 1987) and that these might, for example, be influenced by the specific
 The Voice 161

laryngeal architecture of individual singers (Shipp, Doherty, and Morrissey 1979).

Unfortunately, Shipp et al. do not discuss why their data differ from those of Hirano.
Closer examination of these two studies reveals a number of factors that make it dif-
ficult to compare them in a satisfactory way. These reflect problems that are typical
of the current state of the published literature on the physiological basis of singing.
Both used only four singers and the data obtained are presented in different ways. In
the study by Hirano, the singers were two sopranos, a tenor, and a bass who were all
trained professionals. The study by Shipp used only four males whose vocal range and
proficiency is not reported, though the pitches sung cover the bass to tenor ranges. The
data presented also appear to show that even these four may not all have been using
the same strategy for vocalization. They were also given a number of drugs to reduce
the undeniable discomfort and stress of implanting the electrodes into the laryngeal
muscles, including topical anesthetic to the vocal tract, which has a fundamental effect
on singing performance (see below). Hirano does not describe how his singers were
treated to control discomfort, though this was most probably done. Finally, it was not
clear how the correct placement of the electrodes into the target muscles was ensured.
While these are very difficult experiments to do on human subjects, it is likely that
such approaches will ultimately give considerable insight into singing strategies. It
is therefore a pity that so few have been carried out and that for the time being, the
results of existing studies must be treated with some circumspection.

Nervous Control of the Laryngeal Muscles

In considering the nervous control of the larynx, it is important to remember that in ad-
dition to its role in phonation, the larynx also provides a line of defense for preventing
food material or fluids from entering the trachea. The lining of the larynx has an ex-
tremely abundant sensory innervation, and the purpose of many of the nerve endings
on its inner surfaces is to detect the presence of foreign bodies (Andreatta et al. 2002;
Yoshida et al. 2000). Some of these produce the sensations of pain or discomfort with
which we are all familiar. In addition, however, there are many receptors that gather
information needed by the brain to control the laryngeal muscles during phonation.
This includes feedback from the muscles, joints, and mucosal epithelium within the
larynx as well as from muscles attached to the outside, which adjust its height.
Current theories about how the laryngeal muscles are controlled depend largely on
ideas put forward by Wyke and colleagues, based on a series of studies carried out
in the 1960s and 1970s (Wyke 1974). Only some of the key papers on which these
results are based show raw experimental data, so it is often not possible to assess the
quality of the evidence. Though most of the ideas appear quite reasonable, there have
been few new studies performed in intervening decades using the superior techniques
that are now available. According to the hypotheses presented by Wyke, just before
the onset of vocalization and the accompanying rise in subglottic pressure, there is a
brief burst of activity in the muscles that close the glottis. This would indicate that
the tension, length, position, and shape of the vocal folds is preset prior to any sound
being produced. Most people can quite accurately sing the pitch of a note played on
the piano; however, it has been demonstrated that trained singers generate sounds of
162 Chapter Five

the correct frequency within a few cycles of onset of vocal fold vibration (i.e., within
a very small fraction of a second). As anyone can demonstrate for themselves, this is
under conscious control, but it is too rapid to be a response to hearing the initial pitch
of the sound produced by the voice and then adjusting it to the correct value. Once the
sound is initiated, a considerable amount of sensory information about the vibration
and stretching of the vocal folds and the pressure in the region of the glottis will be
generated as it rapidly opens and closes under the influence of the airstream (Shiba
et al. 1999). Vocal training improves laryngeal control, for example, the ability of a
singer to match a sound whose pitch changes continuously. It is not clear, however,
whether this is primarily a result of an enhanced auditory ability to discriminate pitch
or of a more precise control of the laryngeal muscles in response to sensory feedback.
We are not, in general, conscious of this sensory feedback, though we may feel vi-
brations from other parts of the vocal tract or the tissues in the throat surrounding it.
Wyke’s theory proposes three mechanisms that underlie the control of vocalization
and allow us to match the sound from our own voice to the one we seek to produce.
The first mechanism is a myotactic reflex that causes contraction of the muscles
that close or tense the vocal folds. It is said to be driven by sensory signals from the
laryngeal muscles themselves, generated by the stretching of the vocal folds as they
open and close. The actions of most muscles in the body are monitored by sensory
structures called muscle spindles. In a major review of the control of the larynx writ-
ten by Wyke more than thirty years ago (Wyke and Kirchner 1976) it was stated that
there was considerable controversy concerning whether muscle spindles are present in
human laryngeal muscles. Despite more sophisticated methods of investigation avail-
able today, this controversy remains unresolved (Brandon et al. 2003; Sanders, Han, et
al. 1998; Hirayama et al. 1987; Keene 1961). The one muscle that remains above this
controversy is the interarytenoid muscle that runs between the two arytenoid cartilages.
It clearly does contain muscle spindles (Tellis et al. 2004; Hirayama et al. 1987), sug-
gesting that it may play a key role in the positioning of the vocal folds and stabilization
of the joints between the cricoid and arytenoid cartilages. As we have seen, these are
not simple pivot or ball joints, but allow the arytenoid cartilages to glide forward as
well. Given the degree of freedom this allows, precise control of the position of the
arytenoid cartilages is likely to be central to laryngeal function.
In the absence of muscle spindles, the action of the other intrinsic muscles of the
larynx may therefore be monitored by other types of sensory endings. Amongst these
are so-called spiral endings within the muscle and Golgi tendon organs associated
with the attachment sites of the muscles to the cartilages of the larynx (Wyke and
Kirchner 1976; Nagai 1987). Some physiological studies have found sensory nerve
cells with properties matching what would be expected from such receptors (Shiba et
al. 1999; Sampson and Eyzaguirre 1964), though their presence has not been identified
unequivocally. However, one animal study found that in the thyroarytenoid muscle,
the myotactic reflex is lost following the rather extreme measure of removing the mu-
cosa. The authors interpreted this as implying that there was no sensory contribution
from the muscles themselves (Andreatta et al. 2002); however, it is also possible that
the nerves carrying information from the muscles were damaged by the considerable
disruption that was caused when the mucosa was removed.
 The Voice 163

The second mechanism proposed by Wyke involves receptors in the mucosa of the
larynx that are said to provide sensory feedback that influences muscle activity during
vocalization. There is a particularly high density of these receptors in the region of the
arytenoid cartilages (Davis and Nail 1987; Sanders and Mu 1998). Their physiological
properties have been quite well studied (Shiba et al. 1999; Shiba, Yoshida, and Miura
1995; Shiba et al. 1997; Andreatta et al. 2002; Davis and Nail 1987), but the precise
roles of many of the different types remain unclear (Bradley 2000). Nevertheless, the
basic properties of these receptors, which are found both above and below the glottis,
appear similar to those proposed by Wyke. Some are said to be responsible for the
abrupt rise in the tension in the muscles that close the vocal folds just as the subglot-
tic pressure reaches the level necessary to produce sound (Shiba et al. 1997). With
further increases in subglottic pressure, the tension in the muscles rises more slowly.
Most of the receptors in the mucosa are sensitive to vibrations and some will follow
these accurately up to frequencies of 400 Hz (near G4) (Davis and Nail 1987; Shiba
et al. 1997); however, they are generally most sensitive to lower frequencies. Whether
this frequency information is important for the control of the laryngeal muscles or
whether pitch control is mainly learned through auditory feedback and the memory
stored is not clear. Because they lie close to the surface of the mucosa, the receptors
are rapidly affected by local anesthetics applied to the inner surfaces of the larynx
(Shiba, Yoshida, and Miura 1995). There is some controversy as to the effect this has
on humans. The application of local anesthetic to the mucosa below the vocal folds
has little effect on normal speech, but it was long believed to make singing almost
impossible (Gould and Okamura 1974) because blocking the responses of the subglot-
tal pressure receptors led to a rise in glottal resistance and subglottic pressure (i.e., the
vocal folds were held more tightly closed than normal). However, a more recent study
(Sundberg, Iwarsson, and Billstrom 1995) reported no consistent effect on subglottal
pressure and that singing remains quite possible, though there is a marked deteriora-
tion in pitch control.
The third component of the mechanism proposed by Wyke for the control of
laryngeal muscles is driven by sensory receptors that monitor the position or move-
ments of the joints between the cartilages of the larynx, particularly those linking the
cricoid cartilage with the thyroid and the arytenoid cartilages (Kirchner and Wyke
1964a, 1964b). Some recent findings have been interpreted as suggesting that sensory
information from joints may not after all be involved in the reflex control of muscles
that close the vocal folds (Andreatta et al. 2002). However, this conclusion is again
based on the observation that removing the mucosa abolishes the responses attributed
to joint receptors, and even the authors did not completely dismiss a possible role for
joint receptors in laryngeal control circuitry.
From the foregoing account it will be appreciated that the way in which the larynx
is controlled is still uncertain and that recent studies have complicated rather than
clarified our understanding. Most of our knowledge of the control of the larynx is
based on studies of animals. The sophistication of their vocal repertoire is much less
than that of humans, and though the principles under which they operate are likely to
show many common features, we should expect that the systems controlling human
vocalization will be considerably more complex and precise.
Figure 5-9. The manipulation of formants in voice projection. A, B. Soprano voices. In these dia-
grams, the harmonics of the nonprojected voice are indicated by the dashed lines. When these do
not coincide with the harmonics of the voice source (vocal folds), projection is weak. This can be
improved by manipulating the throat and mouth opening to tune the lowest formant toward the
first harmonic of the note (solid line). As the pitch of the note rises (A to B), the lowest formant
tracks the first harmonic (e.g., by opening the mouth wider) to maintain vocal projection. C, D.
Low voices use a phenomenon called the “singer’s formant” in voice projection. A peak in the
intensity of the sound spectrum of the sung note is produced at 2.5 and 3.0 kHz by the cluster-
ing of several formants. In C, which was recorded from a mezzo-soprano singing a note with a
fundamental frequency of around 300 Hz, the increase in the amplitude of the harmonics due to
the singer’s formant is clearly evident (arrow—see also video 5-2). D. In this frequency range, the
average intensity of the sound from the orchestra is on the wane and so the voice can be heard
above it. This peak will not generally correspond to the pitch of the note, but to some of its upper
harmonics. As will be seen in chapter 9, the brain can nevertheless compute the pitch of the fun-
damental from its harmonic structure. D is reproduced with permission from Sundberg (1972).
 The Voice 165

The Resonator
As already mentioned, the pharynx and the mouth represent the resonator for the
voice and are largely responsible for the harmonic richness that we hear. As with any
wind instrument, the vocal tract resonates best at particular frequencies that depend
on its length and its shape. These resonant frequencies are called formants, and ef-
ficient projection of the voice depends on matching these formants to the frequencies
of the notes being sung (fig. 5-9). A particularly clear discussion of the significance
of formants in singing is given by Sundberg (1977). The frequencies of the formants
set some limits to the effective singing range of individual performers. Experienced
teachers can often predict the likely range of untrained singers by identifying the
formant frequencies of sung vowels, even before the full tessitura of the voice (which
will depend on the structure and muscles of the larynx) has been fully developed. The
pitch of the formants of the vocal tract can, within limits, be manipulated by changing
its dimensions. The formants are derived from standing waves that have a peak ampli-
tude at the vocal folds and a minimum near the mouth opening. If the vocal tract were
a straight tube with a typical length of 17.5 centimeters, the first four formants would
occur at about 500, 1500, 2,500, and 3,500 Hz, which are equivalent to one-quarter,
three-quarters, one and one-quarter, and one and three-quarters of a wavelength.
There is an inverse relationship between the formant frequencies and the length
of the vocal tract and this is reflected in the characteristics of voices with different
ranges, and of voices of the same range but different qualities. For example, the aver-
age formant frequencies found in bass singers are about 15 percent lower than those
in tenors, which is thought to reflect a corresponding difference in vocal tract length
(Titze 2000). In singers with the same range, darker voices may reflect longer vocal
tracts than brighter voices. The effective length of the vocal tract in any one individual
can be altered by about 10 percent by changing the position of the larynx and by ex-
tending or retracting the lips. When the larynx is raised, the pharynx becomes shorter
and its walls tend to sag inward. This leads first to the narrowing and then the closing
off of a cavity that surrounds the upper end of the laryngeal opening called the piriform
recess (Yanagisawa et al. 1991; Benninger et al. 1989). Raising the larynx increases
formant frequencies, which will enhance the upper harmonics, perhaps at the expense
of the fundamental frequency (Sundberg 1999), making the voice sound brighter or
more shrill. When the larynx is lowered, the opposite effect is achieved. The walls of
the pharynx become stretched and its diameter increases, particularly at the base. This
reduces the frequencies of the formants, which results in the lower harmonics of the
voice being more efficiently projected, leading to a darker, more covered quality in the
voice. Many teachers encourage singers to try to maintain a laryngeal position that is
lower than the rest position seen during normal respiration; however, in many, though
not all, trained singers there remains a tendency for the larynx to move upward with
rising pitch, particularly at the upper end of the range (Shipp 1987; Sonninen, Hurme,
and Laukkanen 1999; Yanagisawa et al. 1991). The shape of the vocal tract is not
determined by the position of the larynx alone, however, but by its position relative to
other mobile structures in the neck such as the hyoid bone (Hurme and Sonninen 1995).
We have seen that the larynx can be lowered by the action of the sternothyroid muscles
and raised by the thyrohyoid muscles. However, the actions of the sternohyoid and
166 Chapter Five

omohyoid may also indirectly affect its position and its relationship to the hyoid. Direct
radiographic observations of singers carrying out different vocal tasks reveal complex
and highly variable pitch-dependent changes to the relative position of the larynx and
hyoid and to the shape of the pharynx, whose significance is not currently well under-
stood (Sonninen, Hurme, and Laukkanen 1999; Hurme and Sonninen 1995).
The position of the larynx may also be affected indirectly by how the singer
breathes. When breathing in, the diaphragm contracts and pulls downward, drawing
the base of the lungs with it. It has been suggested that this may put tension on the
trachea and draw the larynx downward (Iwarsson and Sundberg 1998), especially if
the abdomen is allowed to bulge out (the “pear-shape up” posture—see chapter 4).
However, some singers hold the wall of the abdomen in (the “pear-shape down” pos-
ture), which should theoretically limit the downward movement of the diaphragm and
its pull on the larynx. Studies of what happens in practice with these two breathing
strategies reveal a more complicated situation (Iwarsson 2001). When singers prevent
the abdomen from bulging outward, they tend to tilt the head forward, rotating the
chin down toward the neck, which has the result of allowing the larynx to fall lower
in the situation where otherwise it might be predicted to remain high. By contrast, any
tendency to tilt the chin up will increase the upward pull on the larynx, making it rise.
Consequently, when considering the control of the height of the larynx, it is neces-
sary to take into consideration not only the role of the muscles attached to it, but also
posture and breathing technique.
Up until now, we have considered the vocal tract as being analogous to the straight-
ened tube of a trumpet or trombone, but this is too simplistic a notion, as it bends
sharply forward at the back of the throat and its width is far from uniform (fig. 5-1).
Furthermore, its internal diameter can be varied at a number of levels, for example,
by exposing or occluding the piriform recesses (a pair of shallow bowl-shaped spaces
lying on either side of the larynx), by contracting the stylopharyngeus muscle (fig. 5-
2), by altering the position and shape of the tongue so that the mouth cavity becomes
narrowed at different points, or by changing the size and shape of the mouth opening
itself. In the male voice, such means can be used to place the first formant anywhere
between 250 and 700 Hz, and the second formant between 700 and 2,500 Hz (Sund-
berg 1977). For example, the frequency of the first formant is raised when the jaw is
dropped to widen the mouth opening. The second formant can be altered by moving
the body of the tongue and the fourth formant by moving its tip.
Singers with high voices (sopranos) can project the voice adequately when the fun-
damental frequency of the note being sung lies below that of the first formant, but in
the upper part of the range it will generally rise above it. The intensity of the sound
then falls progressively as its pitch diverges more and more from the first formant. To
correct this situation, the singer alters the shape of the pharynx, for example, by open-
ing the mouth aperture to allow the first formant to track the fundamental frequency of
the sung note (fig. 5-9A, B) (Joliveau, Smith, and Wolfe 2004). Constantly moving the
formant frequency in this way not only helps to ensure that the notes are all of equal
volume, but minimizes the vocal effort required by eliminating the damping effect that
the pharynx imposes at pitches where it does not resonate well.
 The Voice 167

In singers with lower voices, the sung note does not generally exceed the frequency
of the first formant, and another mechanism is used for voice projection. When per-
forming with an orchestra, a major problem that singers must overcome is to project
the sound of the voice over it. The sound intensity of the orchestra typically peaks at
between 400 and 500 Hz, and below 100 Hz in particular will quite readily mask the
singer (fig. 5-9C). In classically trained mezzo-sopranos, baritones, and basses, one
noticeable feature of sung vowels is that there is a peak in sound intensity between
2.5 and 3.0 kHz, which is not found in speech (fig. 5-9B, video 5-2). This is called the
singer’s formant (or ring) and is not present to the same extent in sopranos (Sundberg
1999). It appears to be due to a clustering of the third to fifth formants and falls in a
frequency range in which the mean sound energy of the orchestra is declining (fig. 5-
9D). Of course, this does not represent the fundamental of the note being sung and so
does not match its pitch. The perception of pitch is dealt with in detail in chapter 9, but
here it suffices to say that under these circumstances the central nervous system may
use a series of upper harmonics to compute the actual pitch of the voice even though
the fundamental is not clearly audible above the orchestra. There is also another way
in which the singer’s formant may contribute to the audibility of the voice, that the
higher frequencies typical of this range are more efficiently radiated forward (i.e., to-
ward the audience) than frequencies of the lower harmonics of the voice. The singer’s
formant is achieved by widening the pharynx and lowering the larynx (Sundberg
1987), actions that are typical of many male singers (Shipp and Izdebski 1975). As
the pitch of the voice rises, the upper opening of the larynx where it joins the pharynx
may increase considerably in diameter. This changes the sound-projecting properties
of the vocal tract and may compromise the formation of the singer’s formant. To avoid
this eventuality, the larynx must be lowered still farther. This is the opposite of what
happens in normal speech, where the larynx tends to rise as the pitch of the voice in-
creases, a tendency that the singer must therefore learn to counteract (Sundberg 1987).
The lowering of the larynx does not require much muscular effort, so the consider-
able increase in projection that results is still obtained with little extra expenditure of
energy. The importance of the singer’s formant may depend on the context, however.
There appears to be a greater enhancement in this range in solo singing, while in cho-
ral singing more power is generated at the fundamental frequency (Rossing, Sundberg,
and Ternstrom 1986). Boosting power at the fundamental frequency is thought to be
achieved by modification of the voice source, that is, by manipulating the larynx. This
would help to enhance the audibility of the bass parts in choral singing.
The disadvantage of adjusting formant frequency to increase projection of the
soprano voice is that it alters the characteristic of the vowels, and both intelligibility
and some of the expressiveness and inflection present in everyday speech must be
sacrificed in the interest of singing efficiency. However, in lower voices, production
of the singer’s formant has little effect on vowel sounds. To understand the reason
for this effect, it is necessary to consider what gives different vowels their particular
characteristics. This depends on how the properties of the sound source (the harmonic
series generated by the larynx) are modified by the acoustical properties of the phar-
ynx (i.e., its formants). In normal speech, the frequencies of the two lowest formants
Figure 5-10
 The Voice 169

are crucial in enabling us to distinguish between different vowel sounds (Titze 2000).
This can clearly be seen in figure 5-10A, which plots the frequency of these two for-
mants against one another (Peterson and Barney 1952). Figure 5-10B demonstrates
how changing the shape and position of the tongue is used to control the formant fre-
quencies that distinguish [α] (as in car) from [u] (as in boot) and [i] (as in tree). From
this it will be apparent that any manipulation of the vocal tract that alters the position
of these formants to improve voice projection may have the additional consequence of
transforming one vowel into another. Some rules for this are as follows. Constricting
the mouth opening lowers the first formant and raises the second, transforming the
sound toward vowels in which the first two formants are relatively far apart, such as
[i] (as in tree) and [ε] (as in bet). Opening the mouth or constricting the pharynx at
the other end of the vocal tract brings the formants closer together by raising the first
and lowering the second, which pushes the sound toward [α] (as in car) and [o] (as in
bought). As the soprano alters the shape of the vocal tract to optimize the projected
voice as it rises ever higher, the first and second formants of the different vowels
converge and so they progressively sound more and more similar (fig. 5-11). Above
F4 (698 Hz), all vowels in a classically trained projected voice become essentially
identical, though placing the vowel between two consonants does improve matters to
some extent (Smith and Scott 1980). This phenomenon is well understood by com-
posers, who do not score words that must be understood in the upper soprano register
but in this range use the voice purely as an instrument. Interestingly, the question of
vowel intelligibility is rarely raised in the contemporary debates over whether surtitles
should be used for opera sung in the language of the audience.
In singing styles, such as folk or pop, where the intelligibility of the words and
their expression can be particularly important, a more flexible approach to voice pro-
duction is needed than can be achieved in classical singing styles. This depends on a
technique that is more akin to speech and which would make it difficult to sing at high
volume. If the performance takes place in a small venue using acoustic instruments
there is little problem, but where the venue is large or the accompaniment is electroni-
cally enhanced, it may be necessary to use a microphone to maintain the appropriate
balance between the instrumental accompaniment and the voice. The same approach
may be taken in musicals. In nineteenth-century operettas, such as those of Gilbert
and Sullivan, where intelligibility is at a premium and the option of amplification
was not originally available, the more complex soprano arias were scored low in the
vocal range, backed by only light orchestration, and many male arias are sung in a
speechlike declamatory style. In opera, recitative or even speech may be used to take
the story forward.

Figure 5.10. The formant theory of vowel production. The sounds of different vowels can be
characterized by the frequencies of the first and second formants, which are controlled largely by
the placement of the tongue. A. The relationship between vowel sounds and the first two formant
frequencies. The sounds [i], [], and [u] are known as the corner vowels because they represent
the extremes of the vowel space on the graph and also represent the extreme positions of the
tongue. B. The tongue positions for the corner vowels together with the frequencies of the first
two formants. The broken lines at 500 and 1500 Hz indicate the position of the formants with
e
the tongue in a neutral position, represented by the vowel [ ]. Reproduced with permission from
Titze (2000) and Peterson and Barney (1952).
170 Chapter Five

Figure 5-11. Changes in the formant frequencies of different vowels brought

about by the tuning of the soprano vocal tract to optimize voice projection. To
project the voice efficiently, the singer alters the first formant frequency to track
the pitch of the fundamental of the note being sung, for example, by opening the
mouth wider. This causes the frequencies of both the first and second formants of
different vowels to converge and ultimately become identical, making it impossible
to distinguish between them. Reproduced with permission from Sundberg (1972).

The most extreme examples of formant manipulation are encountered among the
practitioners of various forms of throat singing (Levin and Edgerton 1999). This is a
traditional form of vocalization found in areas of central Asia such as Siberia (e.g.,
in Tuva), Mongolia, and Tibet. A single drone note, often with Strohbass qualities,
is sung, sometimes with the involvement of the false vocal folds. The formants of
the vocal tract are tuned to a very narrow peak by manipulating the tongue and lips
to isolate single harmonics of the drone (typically the sixth and above). By picking
these out in a sequence using formant tuning, a melody can be created. Some of the
harmonics are out of tune and so must be avoided. By reducing the opening time of
the vocal folds during each oscillatory cycle, more energy is transferred to the upper
harmonics and sound leakage back into the trachea is reduced, which strengthens the
resonance. There are several styles of throat singing, whose qualities vary from the
gruff to the ethereal.
 The Voice 171

Figure 5-12. Diagrams of the vocal tract during singing the consonant m and the vowels e, o,
and u. This shows the different position of the elements of the resonator used in producing these
sounds. For m, the soft palate is allowed to descend, but for the vowels it seals off the nasal cavity
(this is different from speech). Note also the different shape of the tongue and degree of opening
of the lips in each case. Adapted from Dana (2000).

The Articulator
We take the role of the tongue and lips in the production of the consonants com-
pletely for granted, but with a little experimentation and self-observation, how this
is achieved becomes quite obvious. Figure 5-12 shows drawings based on magnetic
resonance scans of the head that reveal the different position of the lips, tongue, and
soft palate in the generation of different sounds (Dana 2000). The figure shows the
soft palate lowered for the consonant n, and the same is true for m. The sound comes
partly or entirely from the nose. For all of the vowels, the soft palate is raised, block-
ing off the nasopharynx; however, for each the tongue is held in a characteristic way.
172 Chapter Five

This has a dramatic effect on the shape of the oropharynx, and we have already seen
its consequences for formant frequency and the sound of individual vowels.
Some languages lend themselves better to the meter and sound-generating require-
ments of singing; for example, Italian with its many terminal vowels works well while
in French, it may be necessary to distort the language, for example, by sounding the
letter e at the ends of words where it is normally silent in speech. What is more, not
all languages have the same repertoire of sounds and inflections; some are completely
unique to a particular language or group of languages while others, which at first seem
to be similar, may be made in subtly different ways. As a result, learning to speak or
sing in another language often requires the learning of new motor patterns for con-
trolling the elements of the articulator. There is, however, a more insidious problem,
which is that by the time we reach adulthood, we may have lost the ability to identify
or distinguish aurally between certain types of sounds that are not found in our na-
tive language. Without this ability, producing the sounds in speech or singing with
complete authenticity becomes virtually impossible. This is one reason why someone
may be completely fluent in a nonnative language, yet still persist in speaking it with
an accent that is strongly foreign to native speakers, reflecting incorrectly produced
vowels and consonants.

VOICE PROBLEMS

Singing, particularly if this requires strong projection, makes considerable demands

on the physical integrity of the functional elements of the larynx. The vocal folds
and their epithelial covering are delicate structures that are vulnerable to hard usage.
Direct examination of the folds in a number of studies of singers has revealed a high
incidence (60 to 95 percent) of abnormalities, most of which were not at the time as-
sociated with voice dysfunction (Lundy et al. 1999; Elias et al. 1997; Heman-Ackah,
Dean, and Sataloff 2002). While it is likely that many of these will be a direct con-
sequence of singing itself, others may be a reflection of general health problems that
may or may not be obvious to the individual. Stress, poor diet, overindulgence in food
or alcohol, and lack of exercise or rest may all contribute to problems with the voice.
However, there have so far been no comparisons between matched groups of singers
and nonsingers, which might help to clarify these issues.

Environment, Lifestyle, and Medication

Some voice problems arise from environmental conditions. Living or working in an
atmosphere that is dry, dusty, or smoky will almost certainly have consequences for
the voice. Breathing through the nose, rather than the mouth, ensures that the incom-
ing air is warmed and humidified. The inner walls of the nose carry a set of curved
projecting shelves called conchae or turbinate bones, which increase their surface
area. These and the other surfaces within the nose are covered with a moist epithe-
lium that humidifies the air passing through it as well as trapping dust particles. The
epithelium is well supplied with blood vessels, which also warm the incoming air. In
 The Voice 173

a continental climate, at temperate latitudes, the humidity of the atmosphere can at

times be very low, making humidification a much more important issue than it is in
more maritime environments. This is particularly acute in winter during prolonged
periods of low temperature. Habitually sleeping with the mouth open may contribute
to dryness of the vocal tract, and any airborne pollution will also affect the vocal folds.
Smoking generates a particularly acute form of irritation due to the high concentra-
tions of smoke particles that are inhaled. This not only irritates the epithelium of the
vocal folds; the chemicals in tobacco smoke can also damage the structure and the
properties of its cells. This will cause changes to the voice, and may ultimately lead
to the appearance of cancerous growths.
Anything that changes the physical properties of the vocal folds will alter the
characteristics of the voice. A number of factors can lead to an accumulation of fluid
in the tissue of the folds (edema); this may make the voice sound muffled or reduce
its vocal range (Sataloff 1995). The fluid often accumulates, particularly within the
lamina propria, which lies between the mucosal epithelium and the muscular core
of the folds. This region is sometimes known as Reinke’s space and, consequently,
the fluid retention, Reinke’s edema. In extreme cases, the edge of the fold becomes
extended and floppy, giving it a characteristic “elephant’s ear” appearance (video 5-
2). This has a marked effect on voice quality. Paradoxical as it may seem, edema can
occur simultaneously with a drying of the surface of the folds.
In women, mild edema may occur for a few days before menses, during pregnancy,
or, in a small proportion, as a result of taking oral contraceptives. Common medica-
tions such as antihistamines or aspirin may also have effects on the larynx. Antihis-
tamines reduce the secretions from the lining of the nose and throat and may lead to
dryness, while one effect of aspirin is to reduce the ability of blood to clot, so any
hemorrhaging (bleeding) of the vocal folds may be aggravated. Even elements of the
diet can cause vocal problems. In some people, for example, milk products or fats
cause changes in the mucus secretions of the nose and throat, making them thicker and
more troublesome after eating.

Laryngitis
This is one of the most common vocal disorders. It is caused by an inflammation of
tissues in the larynx and is most often the result either of an infection or of gastro-
esophageal reflux (heartburn—see below). Infection of the respiratory tract by bacteria
or more commonly by viruses (e.g., during a cold infection) results in a swelling of the
lining of the larynx due to fluid retention (edema), which is part of the normal response
of the body to infection. This causes hoarseness as the thickening of the vocal folds that
results restricts their freedom to vibrate. Normal cold treatments together with resting
of the voice are in order, and though corticosteroids are sometimes employed to counter
severe swelling, their frequent use is not recommended. In fact, inhaled steroids and
anesthetic sprays seldom help and may actually cause harm, particularly if they are
used to allow performance to continue before the infection has been resolved. Though
steroids may reduce inflammation of the vocal folds, the blood vessels within them
remain fragile; singing under these conditions is associated with an increased risk of
174 Chapter Five

bleeding, which may have both short- and long-term consequences for the voice (Han-
son and Jiang 1998). We have seen that the precise control of the larynx is dependent
on feedback from sensory nerves in the vocal folds; anesthetic sprays used to block the
pain of a sore throat will equally affect nerves carrying information that is needed to
control the pitch of the voice and therefore be avoided if the singer absolutely must per-
form. During laryngitis, it is important to rest the voice as much as possible (even from
talking), particularly at the height of the infection when heavy voice use may result in
long-lasting problems (Koufman 1992). Many singers believe that even whispering
should be avoided in these circumstances. The major reason for this is that though the
vocal folds are held predominantly open, the edges may make periodic contact during
whispered speech. Because the tissue is inflamed in laryngitis, it is particularly vulner-
able to abrasion. The likelihood of vocal fold contact during whispering appears to be
greater in individuals in whom the folds show an outward kinking during this form
of speech than in those in whom the vocal folds are held straight, so it may not be an
equally risky activity in all singers (Solomon et al. 1989). The air turbulence generated
in the larynx during whispering may also tend to dry out the mucosa, which will not
be beneficial to recovery.

Gastroesophageal Reflux (Heartburn)

Laryngitis is sometimes the result of damage to the mucosal epithelium caused when
the acid contents of the stomach rise up into the esophagus and reach the larynx. The
lining of the stomach is resistant to attack by the acid, which is produced by glands in
its walls as a normal function of digestion, but the lining of the esophagus and larynx is
not so protected. When the acid enters the esophagus, it is perceived as heartburn due
to the irritation it causes to the epithelium. Singers with this condition do not always
experience heartburn, but symptoms such as occasional hoarseness not associated with
infection, a feeling of a lump in the throat, difficulty in swallowing, halitosis, or a bitter
taste in the mouth may all be a consequence of acid reflux. The symptoms are often
experienced on first waking. Frequent throat clearing to remove excess mucus, which
is produced in order to protect the pharynx from the acid, may also be a symptom.
Many aspects of professional singers’ lifestyles predispose them to gastroesophageal
reflux (Sataloff 1987). Singers generally perform with an empty stomach, as a full one
makes optimal breathing strategies and support difficult and because the redistribution
of blood to the digestive organs induces a feeling of mental and physical lethargy.
Consequently, the main meal is frequently eaten late at night after the performance and
therefore not long before sleeping. This means that the stomach is active and filled with
acidic liquid when the body is lying horizontally in bed, making reflux more likely.
Treatments for gastroesophageal reflux include medications that neutralize or reduce
the secretion of gastric acid. Simple changes in behavior, such as not lying down until
a couple of hours after eating, and raising the head, neck, and upper body with pillows
while sleeping, will also help (Hanson and Jiang 1998).
A number of other vocal fold problems may result from gastroesophageal reflux,
including the development of growths of various types. Reinke’s edema we have
already mentioned. Another is the appearance of granulomas, which are swellings
 The Voice 175

filled with white blood cells that are removing damaged cells or infectious agents that
have entered the affected area. The first stage of treatment for any such secondary
problem is to identify and address the underlying cause, but surgery may occasionally
be appropriate for granulomas and for Reinke’s edema if persistent in the face of this
approach. If this is necessary, great care must be taken to prevent damage to the free
edge of the vocal fold, which will produce scarring that will badly affect the voice
(Sataloff 1998). However, it must be remembered that the characteristically low pitch
and instability of the voice produced by the edema is for some people an integral part
of their vocal identity, in which case radical treatment is inappropriate unless it is
thought to be related to some serious pathology such as cancer (Sataloff 1998). Small
cancerous growths are sometimes seen on the vocal folds, though these are far less
common than most of the other lesions described here.

Vocal Fold Polyps

Polyps are swellings that are typically seen on only one vocal fold. They may arise
directly from the surface of the fold, or be attached to it by a thin stalk, and are often
supplied by a well-developed small blood vessel. Indeed the polyp may first arise as a
result of a hemorrhage (bleeding) from such a vessel. Though benign (noncancerous),
they may, if allowed to grow, ultimately interfere with the movement of the other
vocal fold. In some cases they will disappear spontaneously following rest and mild
steroid treatment but often require surgery.

Figure 5-13. Photographs of vocal folds. A. Normal folds with a smooth free edge. Blood vessels
run mainly parallel to the edge. B. Vocal folds with nodules. Note that there is one on each fold and
that these lie opposite each other. They resemble calluses and are formed where the folds impact
most strongly together. In this photograph the folds are open (abducted) to reveal the rings of the
trachea beneath. In both images the back of the larynx is at the top. Reproduced with the permis-
sion of Clark Rosen, M.D., University of Pittsburgh Voice Center (voicecenter.upmc.com).
176 Chapter Five

Voice Abuse
This rather dramatic term refers to very loud singing or shouting, overuse of pressed
phonation, or forcing the voice beyond its natural range. Problems can also arise as a
result of using a very strong glottal attack at the onset of the note. The damage caused
to the vocal folds is due to the force with which they are brought together under these
conditions. This can cause bleeding, tearing of the mucosal epithelium covering the
folds (ulceration), and the development of granulomas (see above) or of vocal nodules
(fig. 5-13, video 5-2). Vocal nodules are small calluses (thickenings of the epithelium)
that always appear in pairs, opposing each other where the vocal folds impact most
strongly (Koufman 1992). Though they do not always affect voice quality, they are,
together with other effects of vocal abuse (Elias et al. 1997), often associated with
hoarseness (dysphonia). Vocal nodules are more common in rock, jazz, and gospel
singers, who frequently employ singing techniques (such as pressed phonation) that
have similarities to shouting or screaming (Koufman 1993). The treatment of vocal
nodules has excited some debate. As for most medical conditions that are the result
of a particular activity, the best approach is to eliminate the cause. If this is not done,
then even if the nodules disappear following treatment or are surgically removed, they
will return again when the original singing pattern is resumed. Vocal nodules will
generally (though not always) reduce spontaneously, either with voice retraining alone
or, if appropriate, with treatment for gastroesophageal reflux. This is reflected in an
improvement in voice quality (Holmberg et al. 2001). The growth of vocal nodules is
often associated with a forward movement of the arytenoid cartilages. As a result, the
posterior parts of the vocal folds are hidden when viewed from above, and the poste-
rior part of the glottis may fail to close during phonation (Pontes et al. 2002; Koufman
1993). This symptom can be seen to disappear as retraining progresses. The training
may have to continue for several months, but has an almost identical success rate in
terms of restoration of voice quality (70 percent) to surgical intervention (Koufman
and Blalock 1991), which is popular with some practitioners but carries significant
risks. Retraining should therefore always be the treatment of first resort. If this fails,
surgery may be indicated and if carried out by a careful and experienced surgeon can
be highly successful, but the singer must be fully aware of the risks in order to give
informed consent to the operation.

Vocal Misuse
Like vocal abuse, vocal misuse refers to inappropriate use of the voice, which in the
long run is unsustainable. It can also arise from singing outside the natural range
(tessitura) of the voice or by altering its quality deliberately for the purpose of char-
acterization (Koufman and Blalock 1991). The difference between vocal abuse and
vocal misuse is that the effects of the latter are less extreme, but because the causative
behavior is sustainable for longer, it may have become more ingrained and therefore is
harder to correct. The problems may arise from using higher tensions than are neces-
sary in the laryngeal muscles. High muscle tensions may also be used in an attempt
to compensate inappropriately for mild laryngitis when swelling of the laryngeal
tissue makes it more difficult to maintain vibrations of the vocal folds. Good vocal
 The Voice 177

technique involves the optimization of all aspects of phonation: breath support, con-
trol of laryngeal height and muscle activity, voice projection, and articulation. If one
or more of these elements are deficient, a short-term compensation may be possible
through overuse of the laryngeal muscles, however, ultimately this will break down,
and the only recourse is to perfect all aspects of voice control. Singing, like many
other musical activities, requires prolonged, highly controlled, and sometimes forceful
muscle activity. Warm-up routines, in which the muscles are gradually prepared for
heavy use, are very beneficial in preventing unnecessary tension when the full voice
is finally unleashed. Mental preparation is equally a part of this process, as it allows
the techniques required for the coordination of each of the various aspects of voice
production to be refreshed and rehearsed in a nonstressful way.
In terms of voice care, a singer will naturally concentrate mainly on regulating his
or her singing activities, but care of the voice when speaking should not be neglected.
The ambient noise at social events can reach very high levels, particularly in restau-
rants or reception rooms with bare, reverberant walls. Speaking loudly to maintain a
conversation in such places may entail using pressed phonation, which can be damag-
ing to the edges of the vocal folds in just the same way as other forms of voice misuse.
This can be easily forgotten in the relaxed atmosphere of a postperformance meal or
party.
Singers are fortunate to have many sources of practical information on voice care
as well as access to significant numbers of clinicians with experience of the singing
voice. For example, Care of the Professional Voice by Davies and Jahn is a compact
source of useful authoritative information (Davies and Jahn 2004). Those who wish to
go into this in more detail may find it useful to consult Sataloff’s two-volume Vocal
Health and Pedagogy, now available in paperback (Sataloff 2006a, 2006b).

HOW THE VOICE CHANGES THROUGHOUT LIFE

The Young Voice

The two most important factors in determining the pitch of the voice are the length and
the mass of the vocal folds (fig. 5-14A). At birth, the larynx is small, the vocal folds
are short, and the mean pitch of the voice is about 500 Hz (around B4). The larynx
initially grows rapidly so that by the age of three mean pitch has fallen to around 300
Hz (D4), but from this age until puberty, growth is much slower and mean pitch falls
only a further 25–30 Hz. During this period, however, other changes take place. For
example, at birth there is no well-defined vocal ligament; this only becomes distinct
from the rest of the lamina propria at about the age of eight. At this age the voice
typically has a range of up to two and a half octaves. This remains virtually constant
until the midteens, and any attempt to increase range during this period is likely to
be damaging. Voice training of children should therefore be restricted to improving
sound quality and pitch control (Sataloff, Spiegel, and Rosen 1998). The muscles and
the mucosal lining of the larynx are delicate in children and easily damaged. Child
performers should not be expected to sing as frequently as adults, and in shows with
regular performances, child leads should be alternated among several individuals to
Figure 5-14. A. The effect of age on the fundamental frequency of the voice. The decline in
fundamental frequency from childhood to maturity reflects a change in the length of the vibrat-
ing segment of the vocal folds (the vocal ligament). This is plotted for males and females. The
dashed line indicates the ideal curve that would be produced if vocal fold length was the only
factor controlling frequency. The deviations from this are probably due to changes in the struc-
ture and physical characteristics of the folds. Because the data are averages, the abrupt change
in frequency when the voice breaks in adolescent males is smoothed out. In B, the data has been
replotted to show the change in pitch with age. C. The difference in the dimensions of the laryn-
geal cartilages in adult males and females. This is shown in the horizontal section in D. Because
the arytenoid cartilages form proportionately less of the fold in males, the difference in the length
of the vocal ligaments is 60 percent while the difference in the depth of the thyroid cartilage is
only 40 percent. Reproduced with permission from Titze (2000).
 The Voice 179

avoid damaging the voice. The signs of vocal abuse (e.g., hemorrhaging of the vocal
folds, the development of vocal nodules, edema) are seen in children as well as adults
and can arise either as a result of singing or from prolonged habitual screaming (Reilly
1997).
Of course, the most rapid changes in voice quality take place at puberty. This typi-
cally begins between the ages of nine and fifteen and lasts around eighteen months,
though it can take up to twice as long (fig. 5-14A, C). The greatest changes in voice
quality are generally experienced between the ages of twelve and fourteen (Spiegel,
Sataloff, and Emerich 1997). In young children, the larynx is the same size in both
sexes, but during puberty the male larynx grows more rapidly. By the end of puberty
the thyroid cartilage of the larynx is 40 percent greater in the male than it is in the
female (fig. 5-14B, D). It sits prominently in the neck as the Adam’s apple, which
has a sharper prow-shaped front edge than the female larynx. The male vocal folds
increase in length by up to 60 percent. The change in pitch that occurs around puberty
cannot be accounted for solely by the increased length of the vocal folds (fig. 5-14C).
It is also affected by an increase in the mass of the muscular core of the fold and a
thickening of its mucosal covering, both of which are affected by rising levels of male
sex hormones (androgens) such as testosterone. As a result, the mean vocal pitch falls
by about an octave to 130 Hz (C3). In females, the vocal folds exhibit only about half
the increase in length seen in males and mean vocal pitch descends by only two to
three half-steps to about 220 Hz (A3). The maturation of the muscles of the larynx lags
behind the growth of the vocal folds. One consequence of this is a tendency toward
breathiness in the voice of adolescent female singers. This may be due in part to a
weakness in the interarytenoid muscle, which appears to play a key role not only in
holding the vocal folds closed, but also in stabilizing the joint between the arytenoid
and cricoid cartilages (Tellis et al. 2004).
The changes in the voice at puberty are not due solely to the development of the
larynx. As the body grows, the length and width of the pharynx also increase, chang-
ing its resonant properties. In addition, the tonsils are large in children and can block
the passage to the back of the nose, making the voice sound nasal. In most children,
the tonsils are easily visible at the back of the throat but they decrease rapidly in size
during the teenage years and in most adults are barely detectable. The respiratory
muscles also become stronger and the volume of the chest increases postpubertally.
Muscle strength peaks in the early twenties before vocal skills are fully developed,
and the young adult singer must resist the temptation to use brute force to increase
the volume of the voice in order to compensate for inadequacies in vocal projection
techniques, as this can have long-term consequences for vocal health (Spiegel, Satal-
off, and Emerich 1997). Empirical evidence suggests that when the voice is changing
during puberty, adolescents should be discouraged from heavy voice use and that boy
sopranos and altos should not try to maintain these voices for as long as possible in
the face of the hormonal onslaught. Sataloff proposes that a light diet of choral singing
can be safely continued by allowing the male voices to break naturally during a piece
so that each phrase is delivered in the register that is most comfortable (Sataloff et al.
1997). Though some authors have suggested that intensive vocal training can continue
180 Chapter Five

and indeed be beneficial during this period (Blatt 1983), such a view is highly contro-
versial and is not supported by the majority of vocal teachers.
Psychological factors have the ability to influence the voice at many ages, but
emotional turmoil is often particularly acute during adolescence and young adulthood.
Insomnia, anorexia, and bulimia are common consequences of this in today’s world.
Bulimia has a particularly direct effect on the voice, as bringing up food from the
stomach exposes the throat and vocal folds to acids that can damage and ultimately
scar them. A stressful or frantic lifestyle involving poor eating and drinking habits can
also lead to gastroesophageal reflux, with similar consequences (Spiegel, Sataloff, and
Emerich 1997; Tepe et al. 2002).

Hormones and the Female Voice

For young women, the onset of puberty heralds the appearance of the cyclical hor-
monal fluctuations of the menstrual cycle and for some (though not all) this can have
significant consequences for vocal performance. In order to understand how this
comes about, it is first necessary to describe the sequence of events that underlie the
cycle. The menstrual cycle is ultimately controlled by signals that originate in the
brain, which cause the release into the bloodstream of follicle-stimulating hormone
and leutinizing hormone from the pituitary gland. The effects on the body are brought
about by fluctuations in the levels of sex hormones that as a result are released into the
bloodstream. These come mainly from the ovaries, and the most important are estro-
gen and progesterone. As its name suggests (menses is the Latin word for “month”),
the cycle lasts approximately twenty-eight days (though there is some variation be-
tween individuals), and it can be divided into a number of distinct phases (fig. 5-15).
Because menstruation is the most obvious manifestation of the cycle, the day on which
this commences is designated day 1. Menstruation lasts about seven days, during
which time the inner lining (endometrium) of the uterus (womb) is shed. The ovaries
contain a large number of egg cells (ova), each of which lies within a structure known
as a follicle, which is responsible for hormone production. From the end of menses,
the amount of estrogen released by the follicles starts to rise rapidly, stimulating the
regeneration of the endometrial lining of the uterus in preparation for the possible im-
plantation of an egg. The uterine lining must acquire a structure capable of supporting
a fetus. In order to do this, it becomes thicker and develops an extensive new network
of blood vessels. This is the proliferative phase of the cycle and lasts about a week. If
a fertilized egg later implants and develops, the blood vessels of the embryo will grow
close to those in the wall of the uterus to form the placenta, a structure that allows
nutrients and oxygen to diffuse from the maternal to the fetal circulation. Glands also
develop in the uterine lining, producing either mucus or nutrients for the embryo. Dur-
ing the proliferative phase, the glands at the neck of the womb (the cervix) produce a
watery mucus. This phase ends around day 14 with the release of the egg (ovulation)
from the ovary. This marks the transition from the follicular half of the cycle to the
luteal half. The follicle that is left behind continues to produce hormones, but over
several days the levels of estrogen decline before rising once more, while the produc-
tion of progesterone steadily increases. The progesterone causes the uterine lining
 The Voice 181

Figure 5-15. Hormonal fluctuations during the ovarian cycle. A. The ovaries and uterus. A hu-
man egg (ovum) develops in the ovary and is released into the fallopian tube by means of which
it reaches the uterus. If fertilized by sperm, it will implant into the wall of the uterus and develop
into a fetus; otherwise it is expelled through the neck of the womb (cervix) into the vagina. B.
In preparation for implantation, the lining of the womb (the endometrium) changes in structure
over the menstrual cycle under the influence of the hormones estrogen and progesterone, which
are produced by the tissue that surrounded the egg (the follicle). This tissue remains in the ovary
when the egg is released. The onset of menstruation (menses) is designated as day 1 of the men-
strual cycle. At this time, tissue from the endometrium that developed during the previous cycle
is lost. At the end of menses, levels of estrogen start to rise. The endometrium thickens and its
blood supply develops. When estrogen levels peak, ovulation takes place and progesterone levels
start to rise, further preparing the endometrium for implantation. If this does not happen, the
levels of progesterone and estrogen begin to fall as the follicle degenerates until menses is initi-
ated once more.

to develop further in preparation for implantation and the mucus from the cervix to
become thicker. This is the secretory phase of the cycle, and unless an egg is fertil-
ized and implants into the endometrium, it lasts for about a fortnight, after which the
follicle degenerates and ceases to produce hormones.
At puberty, an increase in the level of sex hormones causes other changes in the
body. Estrogens stimulate breast development and the typical female pattern of fat
deposition, particularly on the hips and thighs. Fatty tissue is itself a minor source of
estrogens. However, “male” hormones (androgens) are also present in women and are
secreted from glands (the adrenals) that sit on top of the kidneys. These stimulate the
growth of pubic and armpit hair and control female libido. High levels of androgens,
for example due to adrenal tumors, can cause the vocal pitch of young girls to fall, but
182 Chapter Five

there is no evidence that these hormones normally have a role in the changes in the
female voice at puberty. Rather, this is correlated with the rise in circulating estrogens
(Pedersen et al. 1990).
Armed with this information, we are ready to consider the effects of female sex
hormones on the voice. Hormones are signal molecules and the body tissues on
which they act must contain proteins known as receptors, which will bind to them if
they are to have any effect. If the hormones are the keys, the receptors represent the
locks. When the two come together, this triggers changes in the behavior of the cells
that make up the tissue. As the female larynx contains receptors that can bind estro-
gens, progesterone, or androgens, all of these hormones have the potential to affect
the voice. Not all singers experience obvious vocal changes related to the monthly
fluctuations in hormone levels, but in about one-third it causes some problems with
the voice. The effects of estrogen and progesterone on the larynx are in some ways
similar to the effects that they have on the uterus (Abitbol, Abitbol, and Abitbol
1999). For example, the rise in levels of estrogen prior to ovulation may thicken the
mucosa of the larynx and increase its watery secretions. This can alter voice quality
to a small degree around the time of ovulation, but this is usually minor and causes
few problems. The most troublesome changes in the voice occur in the four to five
days prior to menstruation (see table 5-3). The rising levels of progesterone reduce
the amount of fluid secreted by the laryngeal mucosa. This thickens its consistency
with the result that the surfaces of the vocal folds become drier. At the same time,
there is an increase in the amount of fluid retained within the folds and a fall in the
tone (tension) of the laryngeal muscles. This causes a reduction in the fundamental
frequency of the voice and can lead to a curtailment of its upper range. The small
blood vessels on the upper surface of the folds become distended and are prone to
rupture. It is advisable that singers should avoid using aspirin or ibuprofen to treat
premenstrual pain or discomfort as both inhibit blood clotting and consequently will
worsen the effects of any bleeding that arises within the vocal folds. Painkillers such
as acetaminophen or codeine do not have this effect. Where treatment is required for
premenstrual vocal problems, this may be purely to provide symptomatic relief (e.g.,
anti-inflammatory drugs to reduce edema); however, hormone fluctuations can also

Table 5-3. Physical and Vocal Symptoms of Premenstrual Syndrome

Physical Change Vocal Consequence
Dryness and loss of vocal fold suppleness Reduced amplitude of vocal fold movement
Thickened laryngeal mucus (reduced volume)
Loss of muscle tone in larynx Vocal fatigue
(opening of glottal chink) Reduced vocal power
Breathiness
Enlarged veins on vocal folds Increased chance of bleeding (especially if
taking aspirin)
Edema (water retention) of vocal folds, Loss of upper range of the voice
reducing amplitude of vibration and Loss of upper harmonics
increasing mass Reduced fundamental frequency
 The Voice 183

be eliminated by using oral contraceptives (which may present their own problems,
as described below).

Contraceptives
In the 1960s, investigations into the effects of the first oral contraceptives revealed
that some had adverse effects on the voice, and this led to the recommendation that
they should be avoided by professional singers (Wendler et al. 1995). Many oral con-
traceptives contain two components, one a synthetic estrogen and the other a synthetic
progesterone. These have three main actions: 1) to prevent ovulation, 2) to thicken
the mucus at the entrance to the womb, which makes it difficult for sperm to enter,
and 3) to make the lining of the womb unable to support an implanted embryo. The
problem for singers is that one class of progesterone-like compounds used in some of
these early contraceptive pills (the 19-norsteroids) mimics to some degree the action
of androgens (male hormones) on the larynx. These compounds can potentially mas-
culinize the larynx by inducing it to grow, which causes the average pitch of the voice
to fall slightly. Even when present in the small amounts found in certain modern con-
traceptives that are not expected to affect the voice, 19-norsteroids can cause hoarse-
ness or discomfort in the throat (Wendler et al. 1995). In many pills, however, they are
replaced by synthetic hormones that not only mimic progesterone, but block andro-
genic effects. As a result, they can also reduce other unwelcome actions of androgens
such as greasy skin, acne, and excessive growth of body hair. Studies of women using
low-dose contraceptives that contain such alternatives to the 19-norsteroids strongly
suggest that they do not have adverse effects on the voice (Amir, Kishon-Rabin, and
Muchnik 2002) and may even have beneficial properties, such as reducing jitter and
shimmer (irregularities in frequency and amplitude of the voice, respectively) (Amir
and Kishon-Rabin 2004; Amir et al. 2003). Though the results presented in these
studies agree with what one would predict, it should be noted that they involved only
small numbers of women, none of whom were trained singers. However, a recent
study has examined the effects of a third-generation combined pill (ethinylestradiol/
drospirenone) on the singing voices of a small group of classically trained, semiprofes-
sional singers (La et al. 2007). During the luteal half of the menstrual cycle, hormone
levels were similar in the placebo and the pill group but during the follicular half,
levels of follicle-stimulating hormone, leutinizing hormone, and testosterone were
reduced, and though no change was observed in jitter, shimmer was also reduced. It
was suggested that the effect on shimmer in the pill group might be due to a reduction
in the incidence of menstrually related abdominal cramps, which can interfere with
support, and the absence of vocal fold edema. Nonetheless, as masculinization of the
larynx is largely irreversible, any decision about the use and choice of contraceptives
should be taken under the guidance of a physician with experience both in vocal func-
tion and in the formulation of different contraceptive preparations.

Menopause and Hormone Replacement Therapy

The onset of menopause typically occurs around the age of fifty and can have a sig-
nificant effect on voice quality. At this time, ovulation ends and the ovaries cease to
184 Chapter Five

produce estrogen and progesterone. Though the levels of sex hormones in the blood-
stream decline significantly, they do not fall to zero. At this time, fat cells become
the dominant source of estrogens, though the quantities generated are relatively small.
The raw material for the production of estrogens is androgens, for which the adrenal
glands are the major source, just as they were prior to menopause. Though the quanti-
ties are quite small, the levels of enzymes in fat tissue that are responsible for convert-
ing androgens to estrogens are higher in postmenopausal women (Misso et al. 2005).
The changes in circulating hormone levels have a number of consequences. We have
already seen that fluctuations in hormone levels during the menstrual cycle affect the
secretions of the mucosal lining of the larynx, and their decline at menopause leads
to a thinning of the mucosa and a reduction in the glands that keep it moist (Abitbol,
Abitbol, and Abitbol 1999). This contributes to a reduction in the suppleness of the
vocal folds, which limits their ability to vibrate freely. There is also a reduction in
muscle strength, not only in the larynx, but throughout the body (Greeves et al. 1999),
including the respiratory muscles, and this, in conjunction with the changes in the
mucosa, can lead to a reduction in vocal intensity and range. These symptoms are
accompanied by a decline in the speed of articulation (Meurer et al. 2004). In many
women there may be no significant change in the pitch of the voice, but in others it
may fall (Lindholm et al. 1997; Abitbol, Abitbol, and Abitbol 1999). Whether this
occurs depends on the relative levels of estrogens and androgens circulating in the
blood and is correlated with body type. Abitbol et al. (1999) explain this in terms of
two types of female figures. The Rubenesque figure has considerable subcutaneous
fat, and as the cells that make up this tissue produce estrogens, the ratio of estrogens
to androgens remains high, helping to ensure that the pitch of the voice remains at
its premenopausal level. In women with a thinner Modigliani-type figure, the much
smaller amount of subcutaneous fat means that the balance of circulating hormones
swings in favor of the androgens. These can consequently have a masculinizing effect
on the voice by triggering growth of the larynx and a fall in vocal pitch after meno-
pause. In addition, the change in the hormonal balance may contribute to dryness
of the vocal fold edges (Lindholm et al. 1997; Abitbol, Abitbol, and Abitbol 1999).
Though it does not affect the voice directly, the decline in estrogen causes a reduction
in the retention of calcium by the bones (osteoporosis), which makes them weaker and
more prone to collapse or breakage. One consequence of osteoporosis of the spine is
that the shoulders may become rounded and the body stooped, both of which tend to
reduce lung capacity.
Many of the changes caused by menopause can be prevented or reversed by hor-
mone replacement therapy (HRT), which is frequently used to combat osteoporosis
and other unpleasant symptoms of menopause such as hot flashes, vaginal dryness,
mood changes, and insomnia. It may also reduce the likelihood of conditions such as
cardiovascular and Alzheimer’s disease, though this remains controversial. HRT is
often prescribed for women who have entered menopause prematurely, for example,
as a consequence of surgical removal of the ovaries or of a hysterectomy. Though
where possible the ovaries are left in the body during the latter operation, the loss of
the womb can still lead to early menopause. HRT does carry a small increased risk
of blood clots and certain types of cancer (notably of the breast and womb) and so is
 The Voice 185

not generally recommended for women with a family history of these conditions, but
many women consider that the small risk is outweighed by the benefits in quality of
life. As with contraceptives, the effect of HRT on the voice depends on the precise
formulation prescribed. In general the treatment should aim to return the concentration
of circulating sex hormones to premenopausal levels. Some forms of HRT include
steroids with androgenic properties. These may be offered orally or by injection for
such conditions as endometriosis, fibrocystic breast disease, or the low energy and
decreased libido that may follow hysterectomy. At any age, these androgens can cause
irreversible masculinization of the voice. Over a period of only a few weeks, such
treatment can alter the pitch of the voice to a degree that would abruptly terminate the
career of a professional singer (Baker 1999). In the eventuality that a singer notices
any voice changes with such medication, she should seek medical advice immediately
and in general should avoid this type of treatment where at all possible because of the
permanent nature of the vocal changes.

The Musici
No discussion of the effects of hormones on the voice would be complete without
a consideration of the castrati, or, as they were rather more flatteringly known in
Italy, the Musici. Records of the castrati performing in the churches and great private
houses of Italy first appear in the mid-sixteenth century (Jenkins 1998; Pain 2006).
Around a hundred years later they began to feature in performances of opera, where
they reached the peak of their popularity in the first half of the eighteenth century
(Somerset-Ward 2004). Though they performed all over Europe, the greatest demand
for castrati during this period was in London, where the best performers could com-
mand immense fees. Leopold Mozart reports that in order to secure the services of
the famous castrato Manzuoli for the 1764–1765 London season, it was necessary to
guarantee him a fee of £1,500 in advance before he would even leave Italy and that in
a single benefit concert in London on March 7, 1765, he made a profit of 1,000 guin-
eas (Anderson 1990). Two of the best known castrati who spent time in London were
Carlo Broschi (who went under the stage name of Farinelli) and Gaetano Majorano
(Cafarelli). Cafarelli sang there for Handel during the 1737–1738 season. The four-
teen-year-old Mozart met both of these musici when he visited Italy in 1770; Farinelli
at his estate near Bologna and Cafarelli in Naples, where he had bought himself a
dukedom and built an opulent villa. With the irreverence of youth, Mozart described
Cafarelli as “a very rich castrato who goes around singing in churches to scrape up
a few coins” (Anderson 1990). With the increasing involvement of female singers,
public taste for the singing of the castrati declined toward the end of the eighteenth
century. Rossini and Meyerbeer were among the last major composers who composed
parts for them. Wagner toyed briefly with the idea of writing the role of Klingsor in
Parsifal for a castrato singer called Mustafa, who is regarded as the last of the virtuoso
castrati (Scammell 1987), but in the end discarded the idea.
Despite documentary evidence for castrati in Vatican choirs from 1589 until the
retirement of the last one (Alessandro Moreschi) in 1913, castration for nonmedical
reasons was always against the law of the church but was conveniently ignored, partly
186 Chapter Five

because of the interdiction against female singers in church choirs. Castrato singers
were never popular in France, and between 1807 and 1815, Napoleon briefly effected
a ban on them when he held control of the Papal States, but it was not until Garibaldi’s
forces entered Rome in 1870 and the Catholic Church lost all temporal power that the
law against it was finally enforced.
The potential wealth that successful castrati could earn in the eighteenth century
was a powerful incentive for many poor families to encourage one of their sons to
undergo the operation, though the chances of reaching the heights of the profession,
or even making a good living from it, were relatively small (Somerset-Ward 2004),
as neither the ultimate quality of the voice nor the singer’s dedication for the intense
training that the profession demanded could be predicted. Castration was usually car-
ried out on boys between seven and nine years of age so that when the normal time of
puberty was reached there was little change in the size of the larynx and their voices
remained relatively high. During the teenage years, their bodies grew in height as they
would normally, and indeed for their time they often became taller than their peers
because the absence of male hormones delayed the end of the growth of the long
bones. However, they developed no facial hair and subcutaneous fat was deposited in
the female pattern. Their skin was smooth and their hair fine and abundant, as they
did not suffer from testosterone-dependent male-pattern baldness. Though a few of
the castrati, notably Farinelli, had a voice that reached into the upper soprano range,
most had what we would now regard as mezzo-soprano (e.g., Cafarelli) or alto voices
(Somerset-Ward 2004).
Joachim Quantz heard Farinelli sing in Milan in 1726 and described his voice as
follows: “Farinelli had a penetrating, full, rich, bright and consistent soprano voice,
whose range extended from the A below middle C to the D three octaves above. . . .
His intonation was pure, his trillo beautiful, his breath control remarkable and his throat
very flexible, so that he performed the widest intervals quickly and with the greatest
ease and certainty. Passage work and runs were of no difficulty to him and in the adagio
his ornamentation was very extensive” (Quantz 1754).
A range of this size was highly unusual among the castrati, for whom two octaves
or less was more typical. The castrato voice was said by some to have the clarity and
purity associated with that of a choirboy, but, as it was produced by a body with the
resonant qualities and lung development of an adult, was a great deal more power-
ful (Jenkins 2001). Its properties were enhanced by a formidable training regimen.
Cafarelli’s daily eight-hour schedule included practice of technically difficult pas-
sages, expressive singing, and deportment while performing. Their technical abilities
were described as prodigious. It was reported that Farinelli could hold a single note for
over a minute without taking a breath, and though this may have been an exaggeration,
one contemporary score indicates a single note held for thirty seconds and ending in
a sustained fortissimo (Depalle, Garcia, and Rodet 1995). Though many eighteenth-
century castrati were notorious for their extravagant exhibitions of pyrotechnical col-
oratura singing and elaborate improvisation, many were famed for their mellifluous
and expressive performances.
The last surviving castrato was Alessandro Moreschi, who became a member of the
choir of the Sistine Chapel in 1883 and its conductor in 1889. He continued to sing
 The Voice 187

in the choir until his retirement in 1913. In 1902 and 1904, sound recordists from the
London division of the Gramophone and Typewriter Company visited Rome to record
the voice of the nonagenarian Pope Leo XIII; however, today they are remembered
not for this, but for their recordings of Moreschi (Moreschi 1987). Moreschi was
never one of the great virtuoso castrati, and by 1902 he was already past his prime as
a singer; nevertheless, these recordings, which are available commercially, do give a
hint of the quality of this otherwise lost sound world.

The Aging Voice

If one compares recordings of the same singer at different periods over a long career,
it is immediately apparent that there are significant changes in voice quality over time.
Though some of this will reflect a maturation of technique, it will also partly be due to
physical changes in the body and larynx. At the latter end of a career, these changes
may start to interfere noticeably with performance. Voice tremor, loss of range, vocal
stamina, and breath control, and difficulty in regulating pitch may start to become
evident. Some of these problems arise from changes in the vocal folds, which with
age become thinner and less elastic and their edges less smooth (Sataloff, Spiegel, and
Rosen 1998). Other changes are related to a decline in the efficiency of the muscles
of the larynx and of the respiratory system. The consequences of menopause aside,
some deterioration in the voice is inevitable in both sexes; however, it is possible
to exert considerable influence over the rate at which this happens. One outcome of
recent research in the field of human aging has been the appreciation that the rate of
physical decline varies greatly between individuals; that is, some people age much
more successfully than others. Though partly genetic, it is to a considerable degree
related to lifestyle and attitude. For example, though the muscles gradually grow
weaker, muscle bulk is strongly dependent on exercise regardless of age. If an arm or
leg is immobilized following a fracture, its muscles waste rapidly even in the young.
In older people, muscle strength and resistance to fatigue can be retained to a great
degree through regular exercise. Maintaining a reasonable level of fitness through
activities such as walking, swimming, or sports will help condition the respiratory
muscles required for singing, while vocal exercises can be used to counter newly
arising technical problems such as vocal tremor. Older singers can also use the skills
developed over a professional lifetime to enable them to perform closer to the limit
of their abilities and thus compensate for some loss of facility. In putting into effect
strategies that will allow the extension of a performing career, attitude is all-important.
When young, we tackle technical obstacles confidently because we have an expecta-
tion that we will be able to overcome them. A similar level of self-belief is needed for
a satisfactory outcome in later years. One should not expect necessarily to continue
singing the same repertoire; indeed, opera singers see the changing roles they perform
during their careers as part of a logical progression. For those in whom singing is
recreational, the physical nature of performing can itself be seen as part of an exercise
regimen that will not only provide physical benefit, but through its social dimension
will contribute to mental health. For further information on this subject the reader is
referred to Sataloff et al. (1997) and Sataloff, Spiegel, and Rosen (1998).
188 Chapter Five

REFERENCES

Abitbol, J., P. Abitbol, and B. Abitbol. 1999. Sex hormones and the female voice. J Voice 13
(3):424–46.
Amir, O., T. Biron-Shental, C. Muchnik, and L. Kishon-Rabin. 2003. Do oral contracep-
tives improve vocal quality? Limited trial on low-dose formulations. Obstet Gynecol 101
(4):773–77.
Amir, O., and L. Kishon-Rabin. 2004. Association between birth control pills and voice qual-
ity. Laryngoscope 114 (6):1021–26.
Amir, O., L. Kishon-Rabin, and C. Muchnik. 2002. The effect of oral contraceptives on voice:
Preliminary observations. J Voice 16 (2):267–73.
Anderson, E. 1990. The letters of Mozart and his family. 3rd ed. London: Macmillan.
Andreatta, R. D., E. A. Mann, C. J. Poletto, and C. L. Ludlow. 2002. Mucosal afferents mediate
laryngeal adductor responses in the cat. J Appl Physiol 93 (5):1622–29.
Baker, J. 1999. A report on alterations to the speaking and singing voices of four women fol-
lowing hormonal therapy with virilizing agents. J Voice 13 (4):496–507.
Benninger, M. S., M. A. Carwell, E. M. Finnegan, R. Miller, and H. L. Levine. 1989. Flexible
direct nasopharyngolaryngoscopy in association with vocal pedagogy. Med Prob Perform
Artists 4:163–67.
Blatt, I. M. 1983. Training singing children during the phases of voice mutation. Ann Otol
Rhinol Laryngol 92 (5 Pt 1):462–68.
Bouhuys, A., D. F. Proctor, and J. Mead. 1966. Kinetic aspects of singing. J Appl Physiol 21
(2):483–96.
Bradley, R. M. 2000. Sensory receptors of the larynx. Am J Med 108 (Suppl 4a):47S–50S.
Brandon, C. A., C. Rosen, G. Georgelis, M. J. Horton, M. P. Mooney, and J. J. Sciote. 2003.
Staining of human thyroarytenoid muscle with myosin antibodies reveals some unique ex-
trafusal fibers, but no muscle spindles. J Voice 17 (2):245–54.
Dana, P. 2000. Orofacial problems. In Medical problems of the instrumental musician, ed.
R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Davies, D. G., and A. F. Jahn. 2004. Care of the professional voice. 2nd ed. London: A & C
Black.
Davis, P. J., and B. S. Nail. 1987. Quantitative analysis of laryngeal mechanosensitivity in the
cat and rabbit. J Physiol 388:467–85.
Depalle, P., G. Garcia, and X. Rodet. 1995. The re-creation of the castrato voice, Farinelli’s
voice. Proceedings of the 1995 workshop on applications of single processing to audio and
acoustics, IEEE, New York.
Elias, M. E., R. T. Sataloff, D. C. Rosen, R. J. Heuer, and J. R. Spiegel. 1997. Normal strobo-
videolaryngoscopy: Variability in healthy singers. J Voice 11 (1):104–7.
Gould, W. J., and H. Okamura. 1974. Interrelationships between voice and laryngeal mucosal
reflexes. In Ventilatory and phonatory control systems, ed. B. Wyke. London: Oxford Uni-
versity Press.
Gray, H., and P. L. Williams. 1989. Gray’s anatomy. 37th ed. Edinburgh: C. Livingstone.
Greeves, J. P., N. T. Cable, T. Reilly, and C. Kingsland. 1999. Changes in muscle strength
in women following the menopause: A longitudinal assessment of the efficacy of hormone
replacement therapy. Clin Sci (Lond) 97 (1):79–84.
Han, Y., J. Wang, D. A. Fischman, H. F. Biller, and I. Sanders. 1999. Slow tonic muscle fibers
in the thyroarytenoid muscles of human vocal folds; a possible specialization for speech.
Anat Rec 256 (2):146–57.
 The Voice 189

Hanson, D. G., and J. J. Jiang. 1998. Laryngitis from reflux: Prevention for the performing
singer. Med Prob Perform Artists 13:51–55.
Heman-Ackah, Y. D., C. M. Dean, and R. T. Sataloff. 2002. Strobovideolaryngoscopic find-
ings in singing teachers. J Voice 16 (1):81–86.
Hillel, A. D. 2001. The study of laryngeal muscle activity in normal human subjects and in
patients with laryngeal dystonia using multiple fine-wire electromyography. Laryngoscope
111 (4 Pt 2 Suppl 97):1–47.
Hirano, M., W. Vennard, and J. Ohala. 1970. Regulation of register, pitch and intensity of
voice. An electromyographic investigation of intrinsic laryngeal muscles. Folia Phoniatr
(Basel) 22 (1):1–20.
Hirayama, M., T. Matsui, M. Tachibana, Y. Ibata, and O. Mizukoshi. 1987. An electron mi-
croscopic study of the muscle spindle in the arytenoid muscle of the human larynx. Arch
Otorhinolaryngol 244 (4):249–52.
Hollien, H., and G. P. Moore. 1960. Measurements of the vocal folds during changes in pitch.
J Speech Hear Res 3:157–65.
Holmberg, E. B., R. E. Hillman, B. Hammarberg, M. Sodersten, and P. Doyle. 2001. Efficacy
of a behaviorally based voice therapy protocol for vocal nodules. J Voice 15 (3):395–412.
Hurme, P., and A. Sonninen. 1995. Vertical and saggital position of the larynx in singing.
Proceedings of the XIII International Congress of Phonetic Sciences, ed. K. Elenius and
P. Branderud. http://users.jyu.fi/~hurme/Sthlm.html (accessed April 23, 2008).
Iwarsson, J. 2001. Effects of inhalatory abdominal wall movement on vertical laryngeal posi-
tion during phonation. J Voice 15 (3):384–94.
Iwarsson, J., and J. Sundberg. 1998. Effects of lung volume on vertical larynx position during
phonation. J Voice 12 (2):159–65.
Jenkins, J. 2001. The voice of the castrato. Trans Med Soc Lond 118:37–42.
Jenkins, J. S. 1998. The voice of the castrato. Lancet 351 (9119):1877–80.
Joliveau, E., J. Smith, and J. Wolfe. 2004. Acoustics: Tuning of vocal tract resonance by so-
pranos. Nature 427 (6970):116.
Keene, M. F. 1961. Muscle spindles in human laryngeal muscles. J Anat 95:25–29.
Kirchner, J. A., and B. Wyke. 1964a. Electromyographic analysis of laryngeal articular re-
flexes. Nature 203:1243–45.
———. 1964b. Laryngeal articular reflexes. Nature 202:600.
Koufman, J. A. 1992. Voice disorders. Visible Voice 1:16–19.
———. 1993. Vocal nodules. Visible Voice 2 (4):74–76.
Koufman, J. A., and P. D. Blalock. 1991. Functional voice disorders. Otolaryngol Clin North
Am 24 (5):1059–73.
La, F. M., W. L. Ledger, J. W. Davidson, D. M. Howard, and G. L. Jones. 2007. The effects
of a third generation combined oral contraceptive pill on the classical singing voice. J Voice
21 (6):754–61.
Laukkanen, A. M., R. Takalo, M. Arvonen, and E. Vilkman. 2002. Pitch-synchronous changes
in the anterior cricothyroid space during singing. J Voice 16 (2):182–94.
Levin, T. C., and M. E. Edgerton. 1999. The throat singers of Tuva. Sci Am 281 (3):80–87.
Lindholm, P., E. Vilkman, T. Raudaskoski, E. Suvanto-Luukkonen, and A. Kauppila. 1997.
The effect of postmenopause and postmenopausal HRT on measured voice values and vocal
symptoms. Maturitas 28 (1):47–53.
Lundy, D. S., R. R. Casiano, P. A. Sullivan, S. Roy, J. W. Xue, and J. Evans. 1999. Incidence
of abnormal laryngeal findings in asymptomatic singing students. Otolaryngol Head Neck
Surg 121 (1):69–77.
190 Chapter Five

Meurer, E. M., M. C. Wender, H. von Eye Corleta, and E. Capp. 2004. Phono-articulatory
variations of women in reproductive age and postmenopausal. J Voice 18 (3):369–74.
Misso, M. L., C. Jang, J. Adams, J. Tran, Y. Murata, R. Bell, W. C. Boon, E. R. Simpson, and
S. R. Davis. 2005. Adipose aromatase gene expression is greater in older women and is unaf-
fected by postmenopausal estrogen therapy. Menopause 12 (2):210–15.
Moreschi, A. 1987. The last castrato: Complete Vatican recordings. Wadhurst, UK: Pavilion
Records, Opal CD 9823.
Nagai, T. 1987. Encapsulated nerve structures in the human vocal cord. An electronmicro-
scopic study. Acta Otolaryngol 104 (3–4):363–69.
Pain, S. 2006. Histories: Superstar sopranos. New Scientist 189 (March 25):52.
Pedersen, M. F., S. Moller, S. Krabbe, P. Bennett, and B. Svenstrup. 1990. Fundamental voice
frequency in female puberty measured with electroglottography during continuous speech
as a secondary sex characteristic. A comparison between voice, pubertal stages, oestrogens
and androgens. Int J Pediatr Otorhinolaryngol 20 (1):17–24.
Peterson, G. E., and H. L. Barney. 1952. Control methods used in a study of the vowels. J
Acoust Soc Am 24 (2):175–84.
Pontes, P., L. Kyrillos, M. Behlau, N. De Biase, and A. Pontes. 2002. Vocal nodules and laryn-
geal morphology. J Voice 16 (3):408–14.
Quantz, J. J. 1754. Johann Joachim Quantzens Lebenslauf, von ihm selbst entworfen [The
course of Johann Joachim Quantz’s life as outlined by himself]. In Historisch-kritische
Beiträge zur Aufnahme der Musik [Historical critical essays on the perception of music], ed.
F. W. Marpurg. Berlin: G. A. Lange.
Reilly, J. S. 1997. The “singing-acting” child: The laryngologist’s perspective—1995. J Voice
11 (2):126–29.
Rossing, T. D., J. Sundberg, and S. Ternstrom. 1986. Acoustic comparison of voice use in solo
and choir singing. J Acoust Soc Am 79 (6):1975–81.
Sampson, S., and C. Eyzaguirre. 1964. Some functional characteristics of mechanoreceptors in
the larynx of the cat. J Neurophysiol 27:464–80.
Sanders, I., Y. Han, J. Wang, and H. Biller. 1998. Muscle spindles are concentrated in the su-
perior vocalis subcompartment of the human thyroarytenoid muscle. J Voice 12 (1):7–16.
Sanders, I., and L. Mu. 1998. Anatomy of the human internal superior laryngeal nerve. Anat
Rec 252 (4):646–56.
Sanders, I., S. Rai, Y. Han, and H. F. Biller. 1998. Human vocalis contains distinct superior and
inferior subcompartments: Possible candidates for the two masses of vocal fold vibration.
Ann Otol Rhinol Laryngol 107 (10 Pt 1):826–33.
Sataloff, R. T. 1987. Common diagnoses and treatments in professional voice users. Med Prob
Perform Artists 2:15–20.
———. 1995. G. Paul Moore Lecture. Rational thought: The impact of voice science upon voice
care. J Voice 9 (3):215–34.
———. 1998. Common medical diagnoses and treatments in professional voice users. In Vocal
health and pedagogy, ed. R. T. Sataloff. San Diego: Singular Publishing Group.
———. 2006a. Science and assessment. Vol. 1 of Vocal health and pedagogy. 2nd ed. San
Diego: Plural Publishing.
———. 2006b. Advanced assessment and treatment. Vol. 2 of Vocal health and pedagogy. 2nd
ed. San Diego: Plural Publishing.
Sataloff, R. T., D. C. Rosen, M. Hawkshaw, and J. R. Spiegel. 1997. The aging adult voice. J
Voice 11 (2):156–60.
Sataloff, R. T., J. R. Spiegel, and D. C. Rosen. 1998. The effects of age on the voice. In Vocal
health and pedagogy, ed. R. T. Sataloff. San Diego: Singular Publishing Group.
 The Voice 191

Scammell, E. 1987. Alessandro Moreschi. The last castrato: Complete Vatican recordings
(sleeve notes). Pavilion Records, Opal CD 9823.
Scherer, R. C. 2006. Laryngeal function during phonation. In Vocal health and pedagogy, ed.
R. T. Sataloff. San Diego: Plural Publishing.
Shiba, K., T. Miura, J. Yuza, T. Sakamoto, and Y. Nakajima. 1999. Laryngeal afferent inputs
during vocalization in the cat. Neuroreport 10 (5):987–91.
Shiba, K., K. Yoshida, and T. Miura. 1995. Functional roles of the superior laryngeal nerve af-
ferents in electrically induced vocalization in anesthetized cats. Neurosci Res 22 (1):23–30.
Shiba, K., K. Yoshida, Y. Nakajima, and A. Konno. 1997. Influences of laryngeal afferent
inputs on intralaryngeal muscle activity during vocalization in the cat. Neurosci Res 27
(1):85–92.
Shipp, T. 1987. Vertical laryngeal position: Research findings and application for singers. J
Voice 1 (3):217–19.
Shipp, T., E. T. Doherty, and P. Morrissey. 1979. Predicting vocal frequency from selected
physiologic measures. J Acoust Soc Am 66 (3):678–84.
Shipp, T., and K. Izdebski. 1975. Letter: Vocal frequency and vertical larynx positioning by
singers and nonsingers. J Acoust Soc Am 58 (5):1104–6.
Sjolander, P., and J. Sundberg. 2004. Spectrum effects of subglottal pressure variation in pro-
fessional baritone singers. J Acoust Soc Am 115 (3):1270–73.
Smith, L. A., and B. L. Scott. 1980. Increasing the intelligibility of sung vowels. J Acoust Soc
Am 67 (5):1795–97.
Solomon, N. P., G. N. McCall, M. W. Trosset, and W. C. Gray. 1989. Laryngeal configuration
and constriction during two types of whispering. J Speech Hear Res 32 (1):161–74.
Somerset-Ward, R. 2004. The castrato ascendancy. In Angels and monsters: Male and female
sopranos in the story of opera, 1600–1900. New Haven, Conn.: Yale University Press.
Sonninen, A., P. Hurme, and A. M. Laukkanen. 1999. The external frame function in the
control of pitch, register, and singing mode: Radiographic observations of a female singer.
J Voice 13 (3):319–40.
Spiegel, J. R., R. T. Sataloff, and K. A. Emerich. 1997. The young adult voice. J Voice 11
(2):138–43.
Sundberg, J. 1972. Production and function of the singing formant. In the Report of the 11th
congress of the International Musicological Society, ed. H. Glahn, S. Sorenson, and R. Ryom.
Wilhelm Hansen edition, 679–688.
———. 1977. The acoustics of the singing voice. Sci Am 236 (3):82–91.
———. 1987. The science of the singing voice. DeKalb: Northern Ilinois University Press.
———. 1999. The perception of singing. In The psychology of music, ed. D. Deutsch. San Di-
ego: Academic Press.
Sundberg, J., J. Iwarsson, and A. H. Billstrom. 1995. Significance of mechanoreceptors in the
subglottal mucosa for subglottal pressure control in singers. J Voice 9 (1):20–26.
Tellis, C. M., C. Rosen, A. Thekdi, and J. J. Sciote. 2004. Anatomy and fiber type composition
of human interarytenoid muscle. Ann Otol Rhinol Laryngol 113 (2):97–107.
Tepe, E. S., E. S. Deutsch, Q. Sampson, S. Lawless, J. S. Reilly, and R. T. Sataloff. 2002. A
pilot survey of vocal health in young singers. J Voice 16 (2):244–50.
Titze, I. R. 1994. Principles of voice production. Englewood Cliffs, N.J.: Prentice Hall.
———. 2000. Principles of voice production. 2nd printing. Iowa City, Iowa: National Center
for Voice and Speech.
———. 2008. The human instrument. Sci Am 298 (1):78–85.
Tsunoda, K., Y. Ohta, Y. Soda, S. Niimi, and H. Hirose. 1997. Laryngeal adjustment in whis-
pering magnetic resonance imaging study. Ann Otol Rhinol Laryngol 106 (1):41–43.
192 Chapter Five

Wendler, J., C. Siegert, P. Schelhorn, G. Klinger, S. Gurr, J. Kaufmann, S. Aydinlik, and

T. Braunschweig. 1995. The influence of microgynon and diane-35, two sub-fifty ovulation
inhibitors, on voice function in women. Contraception 52 (6):343–48.
Williams, R. G., and R. Eccles. 1990. A new clinical measure of external laryngeal size which
predicts the fundamental frequency of the larynx. Acta Otolaryngol 110 (1–2):141–48.
Wyke, B. D. 1974. Proceedings: Laryngeal myotatic reflexes and phonation. Folia Phoniatr
(Basel) 26 (4):249–64.
Wyke, B., and J. A. Kirchner. 1976. Neurology of the larynx. In Scientific foundations of oto-
laryngology, ed. R. Hinchcliffe and D. F. N. Harrison. London: Heinemann Medical.
Yanagisawa, E., J. Estill, L. Mambrino, and D. Talkin. 1991. Supraglottic contributions to
pitch raising. Videoendoscopic study with spectroanalysis. Ann Otol Rhinol Laryngol 100
(1):19–30.
Yoshida, Y., Y. Tanaka, M. Hirano, and T. Nakashima. 2000. Sensory innervation of the phar-
ynx and larynx. Am J Med 108 (Suppl 4a):51S–61S.
 Chapter Six

The Embouchure and Wind Playing

In this chapter we will consider not only the embouchure in its narrowest sense (the
muscles surrounding the opening of the mouth), but also structures such as the jaw and
teeth that support it. We will also investigate the involvement of the larynx and the
throat (pharynx) in wind playing. Finally, we will explore a range of problems that can
afflict wind players, either as a result of a failure of the mechanism of the embouchure
itself or the consequence of generating high pressures in the airway.

THE SKULL AND JAW

During embryonic life, the skull is formed from a number of separate bones that fuse
to form a single structure. Here we need only concern ourselves with some aspects
of the basic structure of the face and jaw, which provide the bony attachments of the
muscles that contribute to the embouchure. Most of the region of the skull that makes
up the face below the eyes is formed by a bone called the maxilla. The upper teeth
arise from sockets that develop in the spongy (alveolar) bone on its lower edge. What
are commonly called the cheekbones (the zygomatic bones) form the lateral edge of
the eye socket. Together with a forward-directed spur from the temporal bone, which
provides part of the side wall of the skull, the zygomatic bone forms an arch (the
zygomatic arch) that runs backward as far as the temporomandibular (jaw) joint (fig.
6-1). Through the space that lies between the zygomatic arch and the side of the skull
runs the temporalis muscle, which is attached at its upper end to the side of the skull
and at its lower end, to the inside of the jawbone (the mandible). Another muscle,
the masseter, runs from the outside of the zygomatic arch to the lower corner of the
outer surface of the jaw. Both of these muscles raise the jaw to close the mouth. The
mandible is made up of two bones, one on the left and one on the right, that are fused
in the middle of the chin. On each side it articulates with the skull via the temporo-
mandibular joint. The saddle-shaped surfaces of this joint also allow the jaw to slide
forward (protract) or backward (retract) and to move from side to side. The joint is a
synovial one (see chapter 1), but unusually it contains a sheet of cartilage, called the

193
194 Chapter Six

Figure 6-1. The muscles contributing to the embouchure are labeled according to their actions
on the embouchure. Anatomists and doctors use the Latin names, which can be found in another
version of the figure on the accompanying disc. Here they have been translated into English. The
central component of the embouchure is the purse-string muscle that surrounds the lips. The
other muscles radiate outward from this, and the direction of pull exerted by each is shown in
the central figure (E).

articular disk, which sits between the two articular surfaces. The muscles that cause
these movements are the medial and lateral pterygoids, which lie deep in the cheek
behind the jaw. In order to allow this freedom of movement, the capsule and liga-
ments that surround the joint and hold it in place are relatively lax. As a result, if the
jaw is opened very wide, or if the chin is struck when open, it can become dislocated,
particularly in those individuals who have loose ligaments in the joint capsule. The
forward and backward movement of the jaw is important for adjusting the airstream
in the playing of many wind instruments. The jaws of some people naturally protrude
more than others, so the degree of protraction or retraction required for different play-
ing tasks will vary from individual to individual.
How the head is positioned on the neck influences how wide open the pharynx (or
throat) is held. The head is supported by a number of muscles attached to the base of
the skull (see chapter 5). On each side a sternocleidomastoid muscle runs from just
behind the ear to the junction of the collarbone (clavicle) and the breastbone (ster-
num). Each rotates the head upward and to the opposite side. When they act together
they pull the head forward, flexing the neck. Other small muscles running from the
 The Embouchure and Wind Playing 195

Figure 6-2. The relationship of the lips and teeth in the playing of different wind instruments.
A. Single reed (clarinet). B. Flute, piccolo. C. Double reed (oboe, cor anglais, bassoon). D. Brass
instruments.

base of the skull to the front of the vertebral column also pull the head forward. All
of these muscles may be active to resist the force of the mouthpiece against the lips.
They are opposed by a number of muscles that run up the back of the neck and pull on
the lower surface of the skull. The head is tilted from side to side by muscles that run
like guy ropes from the sides of the upper vertebrae of the neck to the shoulder blade
and to the upper ribs (the scalenes and levator scapuli; see chapter 2). Damage to any
of the major muscles that do this interferes markedly with wind instrument playing
196 Chapter Six

because of the difficulty in maintaining the throat in an optimum open configuration.

Excessive pressure of the mouthpiece on the embouchure will lead to chronic tension
in the muscles, which may cause pain and discomfort.

THE TEETH

In the playing of wind instruments, the front teeth, particularly the incisors, are used
to grip the mouthpiece (woodwind instruments) or support the embouchure (brass
instruments) (fig. 6-2). For single-reed instruments, the reed is gripped between the
upper incisors, which make contact with the top of the mouthpiece, and the lower lip,
which is curled over the lower incisors. For double-reed instruments, both lips are
curled over the incisors to apply the pressure to the reed. The teeth are inserted into
the spongy alveolar bone of the upper and lower jaw. Though this forms a rigid plat-
form in the short term, bone can be remolded over time in response to pressure. This
property allows orthodontists to straighten or adjust tooth position using tooth braces.
To do this, braces are used to generate forces of between 36 and 60 grams over peri-
ods of six hours or more at a time. The pressure applied directly to the teeth by wind
instruments can be much higher. Average pressures of 211 grams have been reported
for the flute, 270 grams for reed instruments, and 500 grams for brass instruments
(Engelman 1965); however, one should treat these values with some caution as they
were made on very young players (ten to seventeen years) with as little as one year’s
experience on their instrument. As we will see later, a study of the pressures applied
to the mouthpiece by the lips of adult trumpeters revealed values four to ten times as
high (Barbenel, Kenny, and Davies 1988). Though it is clear that wind instrument
playing has the potential to affect tooth alignment, studies of this in adult players have
been inconclusive. It may be that the duration of the practice sessions typical of adult
amateur musicians is too short to have a major effect on the teeth, though there is some
evidence for changes in tooth alignment in musicians under the age of fifteen. Indeed,
a suitable choice of instrument can apparently aid orthodontal treatment for those in
this age group (Yeo et al. 2002)! When the surfaces of the molars come together, the
cutting surfaces of the upper front teeth (incisors and canines) overlap slightly over
those of the lower front teeth. The vertical overlap of these edges is called the overbite,
and the horizontal overlap (the margin by which the top teeth project over the bottom
teeth) is called the overjet (fig. 6-3). If the lower incisors overlap the ones on the upper
jaw, this is called underjet or underbite. Single-reed instruments can increase overjet
and overbite, while double-reed instruments tend to reduce overjet and increase over-
bite, and brass instruments reduce both overjet and overbite (Herman 1981).
Irregularities and sharp or chipped edges on the anterior teeth can pose particular
problems for reed players. The contact between the lower incisors and the lip is the
main source of trauma for players of single-reed instruments, while for players of
double-reed instruments, both the upper and lower lips may be affected. In the short
term, tenderness and ulceration may result. Over time, the soft lining of the inside
of the lip will change its characteristics so that the cell layers at the point of contact
acquire more granules of keratin (a structurally strong protein), forming a protective
 The Embouchure and Wind Playing 197

Figure 6-3. A diagram with an explanation of the terminology used for describing the relative
position of the upper and lower incisor teeth.

callus. This quickly develops in all players who practice consistently; however, seri-
ous problems with the lips in professional players can be alleviated by having chips or
sharp edges on the teeth rounded or repaired. A lip shield can also be used to reduce
the discomfort. In single-reed players, the upper incisors may become worn due to
friction against the mouthpiece. The application of rubber patches to the top surface of
the mouthpiece is now common practice, and acrylic shields can be made to cover the
upper incisors and canines in order to protect both natural teeth and vulnerable dental
work such as crowns (Herman 1974).

MUSCLES OF THE MOUTH, JAW, AND EMBOUCHURE

The muscles around the mouth (oral cavity) that are associated with singing and wind
playing can be divided into two groups. The first group contains the deep muscles
that move the jaw or form the inner wall of the cheek and the second, the superficial
muscles that insert into the lips (fig. 6-1).

Deep Muscles
Two large muscles are responsible for closing the mouth. These are the masseter
and temporalis muscles. If you clench your teeth tightly and put your fingers on the
angle of the jaw, a couple of centimeters below the ear, you will feel a tense ball of
muscle. This is the masseter, which links this part of the jaw to the cheekbone (the
zygomatic arch). If you repeat the action of clenching your teeth but this time place
your fingers on the side of the head above the cheekbone but behind the eye, you will
198 Chapter Six

feel the contraction of another large muscle, the temporalis. This descends behind
the zygomatic arch and attaches to a sheet of bone (the coronoid process) that rises
upward from the jaw. The fibers of the temporalis muscle fan out from the coronoid
process so that although those running straight upward close the jaw, others run
backward almost horizontally so that parts of the muscle pull the jaw backward (i.e.,
retract it). The jaw is pulled forward by muscles (the pterygoids) that run from the
back of the lower jaw to a region of the skull behind the upper teeth. The jaw may
be protracted and retracted during playing when changing registers, for example, in
brass instruments such as the trumpet (Ridgeon 1986)—while in the flute, the lower
jaw may be retracted in the lower register.
Deep to the temporalis and masseter is another sheet of muscle that attaches at one
end to the bone above the upper teeth, and at the other end to the outer surface of
the lower jaw below the teeth roots. This muscle is called the buccinator (its name is
derived from the Latin word for a trumpeter). It forms the inner wall of the cheek and
extends forward to the opening of the mouth, where it contributes to the embouchure.
At the back it merges with the outer muscle layer of the pharynx (the superior con-
strictor—see chapter 5) so that the muscular wall of the throat is continuous all the
way to the lips. This ensures that the airway can resist the high pressures generated
in the playing of many wind instruments. In some trumpeters who allow their cheeks
to puff out, the buccinator is stretched and can become ever thinner until it reaches
a state in which it is easily damaged. The muscles that move the jaw are all supplied
by branches of the trigeminal nerve, which is responsible for sensation on the face.
By contrast, the buccinator and the muscles of the embouchure, described below, are
supplied by the facial nerve.

Superficial Muscles
The embouchure, per se, is formed by muscles of the superficial group. Figure 6-1
describes them by their actions, in English. A version of this figure with the muscles
named in Latin can be found on the accompanying disc. The Latin names are the ones
that are used in the reports of studies into embouchure function in wind players. The
superficial muscles are known as muscles of facial expression. They are attached to
the overlying skin, and we use them to give conscious or unconscious indications of
recognition and of mood or to provide signals that reveal how we feel about what is
going on around us. Their Latin names indicate their relationship to the opening of the
mouth (oris) or the lip (labium). The central muscle of the embouchure is one called
the orbicularis oris because its fibers encircle (or orbit) the mouth. Some of its fibers
are attached to the bones of the upper and lower jaw and others to the deep layers of
the skin. The muscle has a purse-string action, which is most graphically demonstrated
in pouting. The other muscles that control the shape of the embouchure radiate out
from the orbicularis oris (to which they are attached) to various anchor points on the
bones of the face and jaw. The upper lip can be raised by a muscle that runs up to the
bone under the eye and on the side of the nose (hence its name—levator labii supe-
rioris). The corner of the mouth is raised by a small, deep triangular muscle (levator
anguli oris) and pulled upward and backward by muscles that run from the zygomatic
 The Embouchure and Wind Playing 199

arch of the cheek (zygomaticus major and minor). A comparable set of muscles act on
the lower lip. These all run from the lower jaw and they can depress the lip (depressor
labii inferioris and mentalis) or the corner of the mouth (depressor anguli oris). The
region at the corner of the mouth where the levator and depressor anguli oris and the
zygomatic muscles all merge with the orbicularis oris is called the modiolis. In addi-
tion to these muscles, a much smaller one, the risorius (smiling muscle), contributes to
the embouchure by pulling the corner of the mouth backward. In this it is aided by the
buccinator and by the platysma. As its name suggests, platysma is a thin, flat sheet of
muscle that runs beneath the skin of the neck, to which it is attached. Its upper fibers
pass over the lower edge of the jawline and interleave with the muscles that contrib-
ute to the embouchure. The branches of the facial nerve, which supply the muscles of
the embouchure, radiate out from under the largest of the salivary glands (the parotid
gland), which lies beneath the skin over the angle of the jaw.

THE EMBOUCHURE IN WIND PLAYING

The role of the embouchure varies depending on the instrument being played, but in
all cases its shape must be precisely regulated using not only the orbicularis oris, but
also the other muscles that surround it. Sometimes when trying new embouchure con-
figurations, fatigue and stiffness may be detectable in individual muscular elements
of the embouchure. In the case of reed instruments the first role of the embouchure
is to maintain a seal around the mouthpiece or reed(s), often against high intraoral
pressures. It is also necessary to apply a variable pressure to the reed depending on
the sound being produced, and this may involve the muscles that raise the lower jaw
(temporalis and masseter) as well as those controlling the lips. In the flute, the airhole
is always open but the precise shape of the embouchure is adjusted to direct the air
jet and control its length. In brass instruments, the basic pursing of the lips is also
achieved by the orbicularis oris. Players of these instruments generate extreme intra-
oral pressures in the upper registers (see chapter 4). The tension in the orbicularis oris
will determine the frequency at which the lips vibrate to generate the sound. This in-
volves not just contraction of the fibers of the orbicularis oris itself, but of the muscles
radiating out from it. Indeed it is the activity of this latter group that dominates in
expert players. In order to gain the degree of control over the embouchure needed to
allow precise control over pitch in all registers, the muscles that contribute to it must
be strengthened over time and be kept in good physical condition by constant training
to maintain good performance standards. In an active player this may noticeably alter
the shape of the face. In the French horn, where the overtones of the instrument used
for the notes in the upper register are very close together, the control of the embou-
chure is particularly critical. For brass players, the precise shape of the embouchure
depends not only on register but also on the instrument, as mouthpieces vary greatly
in size and shape (Ridgeon 1986). It is not just the shape of the mouth aperture that is
used to control the airflow. The position and shape of the tongue also play a crucial
role. For example, it may be raised to reduce the area of the air jet (and presumably
increase its pressure) in the upper register (Ridgeon 1986). In the flute, the opening
200 Chapter Six

of the embouchure itself is reduced as pitch increases, for similar reasons (Cossette,
Sliwinski, and Macklem 2000).
The question of the optimal pattern of muscle activity in the embouchure is cen-
tral to the teaching of brass technique; however, only a few studies have examined
this directly and all have focused on trumpet players (White and Basmajian 1973;
Basmajian and White 1973; Heuser and McNitt-Gray 1991, 1993, 1998). Basmajian
and White recorded the electrical activity in embouchure muscles using small wires
that were inserted directly into them through the skin. They monitored the upper and
lower regions of the orbicularis oris and the levator and depressor anguli oris muscles,
as they considered these to be the most crucial elements of the embouchure for this
instrument. The results demonstrated that despite being considered as a single muscle
from an anatomical standpoint, the fibers of the orbicularis oris in the upper and lower
lip could be controlled separately. Muscle activity was compared in players with a
wide range of expertise ranging from teenage beginners to adult professionals. In all
players, there was greater muscle activity when playing in the upper registers than the
lower, and when playing loudly as opposed to softly. When playing a note, muscle
activity was first seen in the preparatory period before its onset. This more closely
matched the activity required during the note in experienced players than in begin-
ners, a finding confirmed by other investigators (Heuser and McNitt-Gray 1991). This
presumably reflects a greater accuracy in note placement. Basmajian and White found
that in all players, the end of the note coincided precisely with an abrupt and complete
cessation of all muscle activity, though other studies have suggested that this is more
marked in those with most experience (Heuser and McNitt-Gray 1991). In advanced
players, the activity in the muscles surrounding the lips (i.e., levator and depressor
anguli oris) was greater than in the lips themselves (the orbicularis oris) and there
was no difference in activity between the upper and lower lip. In beginners, there was
generally more activity in the upper lip. The players in the study were asked to carry
out a number of simple exercises including small-interval lip slurs and both slurred
and tongued arpeggios and repeated notes. The advanced players showed very little
variation in muscle activity during the small-interval slurs, suggesting that they were
using some other method to manage the transition between the notes. It was proposed
that mouthpiece pressure, pressure distribution on the lips, air pressure, airflow, tooth
aperture, and tongue position might be candidates for this, but the question was not
investigated further. The experienced players also showed much smaller differences in
the pattern of muscle activity used for slurred and tongued arpeggios than the novices.
Electrical recordings from muscles of the embouchure can be made less invasively
using electrodes that are applied to the surface of the skin. Though less accurate than
making recordings via wires inserted into the muscle, this technique does not involve
breaking the skin and so can be used much more easily in a nonclinical context (Heu-
ser and McNitt-Gray 1991). Heuser and McNitt-Gray recorded from the zygomaticus
major and the depressor anguli oris. Surface recording of embouchure muscles may
have a role in normal teaching aimed at correcting or refining embouchure technique.
For example, in some players, poor tone production was associated with a lack of
synchronization of muscle activity within the embouchure or an imbalance in the
activity of equivalent muscles on the left and right sides (Heuser and McNitt-Gray
 The Embouchure and Wind Playing 201

1998). In addition, players with a history of attack difficulties showed poor control of
the onset and offset of muscle activity. Resolution of these difficulties is accompanied
by a clearly observable change in the pattern of electrical activity of the muscles.
Brass teachers generally recommend that the mouthpiece be placed centrally on the
lips because this is thought to be optimal for tone quality; however, some experienced
players do perform with it displaced to one side. This can be due to irregularities in the
upper incisors that make central positioning uncomfortable (Nixon 1963). One might
expect that this would be associated with some asymmetry in embouchure muscle ac-
tivity, but a comparison of three such players with a group that used the more normal
configuration showed no observable differences between them over a range of playing
tasks (Heuser and McNitt-Gray 1993).
The pressures applied to the mouthpiece by lips of trumpet players has been thor-
oughly investigated by Barbenel, Kenny, and Davies (1988), who tested sixty per-
formers, half of whom were top-level professional orchestral, jazz, or band musicians
and half of whom were less-accomplished players with at least six years of playing
experience. For all subjects, mouthpiece pressure rose increasingly more rapidly with
rising pitch (i.e., rose in a nonlinear fashion). For notes in the middle to upper part
of the range there was little difference between the two groups in terms of the forces
used. For example, for the note G5 played at 95 decibels using a normal level of force,
the mean range was 2–2.7 kilograms (kg) (20–27 newtons [N]). When players were
asked to produce notes at the upper extreme of the range (C6–G6), less than half were
successful. Those who were could typically tolerate a mouthpiece pressure of 7.7
kg, while those that were not could support only 4.5 kg. This contradicts the popu-
lar myth that virtuoso performers can play when exerting almost no pressure on the
mouthpiece. Reducing the pressure by as little as 0.3 kg in this study was sufficient to
render it impossible to continue playing a note at the same intensity. A smaller study
of college-level trumpeters confirmed that while similarly high pressures were some-
times used for short periods, the average pressures exerted were relatively low—in
this case about half those measured by Barbenel (Devroop and Chesky 2002). The
perception that proficient or virtuoso players use very little force may stem from the
fact that when it comes to estimating visually how much force is being used, both
professional performers and nonmusicians do rather poorly (Barbenel, Davies, and
Kenny 1986). What watching players does reveal is how much effort they are using to
produce the tone—in other words, how close they are to their physical limits. Hence,
highly proficient players give the illusion of using low mouthpiece pressures because
they are playing well within their capabilities. Interestingly, it also proved difficult for
players of all proficiencies to estimate how much pressure they were applying to the
mouthpiece themselves. Both groups tended to think they were using a greater range
of pressures than they actually were.

THE ROLE OF THE LARYNX AND PHARYNX

Several studies involving direct visualization of the larynx have been directed at the
role of the glottis in wind playing (Rydell et al. 1996; King, Ashby, and Nelson 1989;
202 Chapter Six

King, Ashby, and Nelson 1988; Brown 1976). Though these studies present rather brief
and incomplete descriptions of laryngeal function, some general features were observed
consistently. The larynx appears to be used to control airflow in all wind instruments,
though this varies in degree and in context between instruments. In the flute, the glottis
is kept wide open during playing, but rhythmic movements of the vocal folds were con-
sistently seen to contribute to the production of vibrato. In reed and brass instruments,
the contribution of the glottis to the production of vibrato varies considerably between
players and is sometimes absent entirely. During the playing of these instruments, the
two vocal folds are actively brought into close proximity (though not into contact). In
brass players this is more marked in instruments that require high pressure and low air-
flow, such as the trumpet, than in larger-bore instruments that require higher flow rates
at lower pressure, such as the tuba. It is has been proposed that in the trumpet this may be
a first stage in which fine control may be exerted on the pressure being generated. This
may be aided by a depression of the epiglottis and by use of the tongue. In brass and reed
instruments it was noted that the reduction in the opening of the glottis was most marked
in heavy tonguing, during which the arching of the back of the tongue may have also
contributed to the reduction in airflow. Complete closure of the glottis in this situation
is seen most commonly in the trumpet and reed instruments. Whether this closure was
functional or represented an inappropriate engagement of the valsalva maneuver was not
discussed in any of the above-mentioned studies (see chapter 4).
The height of the larynx may also be manipulated during woodwind playing. A
lowering of the larynx has been noted during playing of the flute, saxophone, and
oboe (Rydell et al. 1996; King, Ashby, and Nelson 1988). In the flute it is maintained
in this low position, and changes in tonal quality are brought about by altering the
embouchure and the direction of the air jet. It has been suggested, however, that
the resonance of the cavity of the pharynx may be used, particularly by advanced
woodwind players, to contribute to the timbre of the sound (King, Ashby, and Nelson
1988; Benade 1986). The resonant properties of the pharynx are affected in part by
the height of the larynx, to which of course it is attached, and are greatest when the
glottis is held nearly closed during playing. When the glottis is opened, the resonances
are weakened by being dissipated into the lungs. In the teaching of brass players, the
instruction to play with an “open throat” refers to the pharynx; playing with a tight
throat is said to produce a pinched tone, suggesting that resonance in the cavity of
the pharynx contributes to the quality of the sound. In the oboe it has been proposed
that a maintained low position of the larynx is necessary for the development of the
harmonic structure typical of this instrument, presumably as a result of the effect this
has of increasing the volume of the pharynx, though the validity of this has not been
tested experimentally. In single-reed instruments, the movement of the larynx varies
considerably between individual players and can be difficult to quantify. Some reports
suggest that it tends to rise with pitch (King, Ashby, and Nelson 1988), while others
state that it remains in a stable position (Weikert and Schlomicher-Thier 1999). It has
been proposed that the diversity in technique may be related to individual variability
in the physical dimensions of the vocal tract.
In the past, technical difficulties in making measurements of the acoustic imped-
ance of the airway during playing had resulted in considerable disagreement concern-
 The Embouchure and Wind Playing 203

ing the importance of vocal tract resonances in woodwind playing; however, recent
advances in measuring equipment have made accurate analysis a more realistic possi-
bility (Fritze and Wolfe 2005). In the case of clarinet players, it appears that the vocal
tract resonances remain fairly constant throughout most of the range; however, they
may be manipulated to a considerable extent in the highest register and for special
effects such as glissandi and slurring. Some players report that they move the tongue
and facial muscles between positions that are similar to those typical of certain vowels
(e.g., “ee” for the top register versus “aw” for a full sound in the low register). Of
course, any change in jaw or lip configuration that accompanies these configurations
may also alter the loading on the reed and so the results require careful interpreta-
tion; however, it at least appears feasible that the magnitude of changes in vocal tract
resonances might be capable of affecting the quality of the sound, and certainly many
players believe this (Fritze and Wolfe 2005).
One instrument in which the major contribution of vocal tract resonances to the
sound produced is unequivocal is the Australian aboriginal didgeridoo (Amir 2004).
This is an unusual instrument both structurally and acoustically. Its bore varies ran-
domly along its length and it has no mouthpiece or narrowing at the playing end. The
fundamental frequency is strong though very low (typically around 70 Hz) where the
auditory system is not particularly sensitive. In good instruments, the lowest harmonics
(typically the second to ninth) are weak or absent. The vocal tract is strongly coupled
(though in a complex way) to the instrument through the lips, which act as the reed. The
frequencies generated from the tenth harmonic upward appear to be dominated by the
formant frequencies of the vocal tract; the sound can be modulated by using vowel-like
conformations of the mouth, while the vocal folds remain open and free of vibration.
This manipulates the relative frequencies of the formants (see chapter 5) and strongly
influences the timbre of the note produced by the instrument.

PROBLEMS ASSOCIATED WITH

THE EMBOUCHURE IN WIND PLAYERS

Dental Problems
Clarinet and saxophone players who have a pronounced overbite may experience
excessive wear on the incisors (Yeo et al. 2002). The overbite results in the frictional
wear being applied to the back of the edge of the tooth. If the wear extends beyond the
outer layers of the tooth, the central pulp, which contains its nerve and blood supply,
can be affected and the tissue may die as a result. A number of dental interventions
can be used to treat this. If the wear has not already affected the pulp, a surface coat-
ing to the back of the tooth will provide useful protection; however, if the pulp dies,
a root canal filling with a post and crown would be necessary to preserve the tooth.
This involves removing the pulp and inserting a rigid post to support the crown, but
the tooth is weakened as a result.
The effect of backward pressure exerted by the mouthpiece on the teeth remains
uncertain. One study reported that in single-reed and brass players the incisors were
204 Chapter Six

inclined backward twice as frequently as in the general population (Gualtieri 1979),

while another noted only minor differences (Rindisbacher et al. 1990). The pressure
on the teeth does, however, produce significant problems for those with false front
teeth. Absence of any natural teeth presents almost insuperable obstacles for anchor-
ing removable dentures sufficiently to support the embouchure, while retention of
even a small number (e.g., two to five) can provide enough support for dentures if
properly designed (Fine 1986). Surgical implanted supports for the dentures can be
effective where there are no teeth to support the dentures. If any front teeth must be
removed, a cast should be made so that the replacement dental prosthesis will match
them as closely as possible in order to minimize the adjustments needed in playing.
Indeed, for professional players with sound teeth, it is a sensible insurance policy to
have casts made to facilitate reconstruction in case of future accidents. Even loss of
the teeth at the back of the mouth can result in difficulty in controlling the airstream
and so their replacement should not be neglected.

High Pressure in the Airway

The possible effects of high pressure in the lungs of wind players is discussed in chap-
ter 4. Though the tissues within the lungs might theoretically be compressed, the chest
cavity is well supported against high internal pressures by the rib cage; however, other
regions of the airway in the pharynx and oral cavity (mouth) do not have skeletal sup-
port and may undergo distension and damage due to high internal pressures. We will
start by considering problems due to weakness of the soft palate. The high pressures
that are required in playing brass and reed instruments can only be supported if the
passageway from the throat to the back of the nose is efficiently blocked off by the rais-
ing of the soft palate against the back of the nasopharynx (see fig. 4-14). Occasionally
this mechanism fails in wind players, a situation that is known as loss of seal or, more
technically, velopharyngeal insufficiency (Ziporyn 1984; Gilbert 1998). The symptoms
are either an inability to generate enough pressure to play certain notes, or noise com-
ing from the nose when certain notes are played. It would be logical to assume that
this is particularly associated with the instruments that require the highest maximum
pharyngeal pressures, for example, oboe, horn, and B-flat and piccolo trumpets when
playing in the upper registers. To some extent this is true, but it is not the whole story
as it does not explain why clarinetists are among the wind players most commonly
affected. Though the maximum pressures generated when playing the clarinet are far
outstripped by the other instruments mentioned above, it does require among the high-
est minimum pressures to initiate notes (Schwab and Schultze-Florey 2004). Put more
simply, to initiate the notes in the lower and middle registers, which is the part of the
range in most constant use, requires higher pressures than notes in the equivalent range
of the horn and trumpet. As a result, the muscles of the soft palate in clarinetists must
on average do more work than those in players of small-bore brass instruments. Though
not the sole cause of loss of seal, fatigue in the muscles of the soft palate (the tensor
and levator palatini muscles) is the major one. The problem frequently arises as a result
of a period of intense playing maintained over a number of days or weeks. It may also
occur following a break from playing (such as a holiday), because the muscles have
 The Embouchure and Wind Playing 205

lost some of their condition, or be experienced during a cold infection or during a time
of emotional stress (Schwab and Schultze-Florey 2004). Loss of seal is most common
in young players (particularly in females), either because the muscles of the soft palate
are not fully developed or, less often, because the soft palate as a whole is too short to
make unobstructed contact with the back of the throat. Depending on circumstances,
complete rest or a return to less-demanding repertoire, followed by a program of pro-
gressive exercise for the muscles of the palate, is generally sufficient to resolve the
problem. It can also arise following tonsillectomy, after which it is not unusual for wind
players to experience difficulties for three to six weeks or more. During this period,
great care should be taken not to damage muscle and other tissues around the site of the
tonsils (Hart and Logemann 1986). The presence of enlarged adenoids (tonsils), which
is common in children and teenagers, may in itself prevent the formation of an efficient
seal. If loss of seal for whatever reason is persistent in professional players, a prosthesis
(a physical support), typically a plastic plate similar to one used to hold dentures but
extending backward beyond the hard palate, is one option to keep the soft palate in
position during playing. This is not worn constantly but is slipped in only for playing.
Alternatively, if this fails or interferes with playing, it may be possible to enlarge the
soft palate by surgical means (Dibbell, Ewanowski, and Carter 1979). The least major
form of surgery is to inject some inert material under the surface of the soft palate
where it fails to make adequate contact with the back of the throat. This material can be
fat taken from the player’s own body (Klotz et al. 2001), or a man-made material such
as Teflon (Gordon, Astrachan, and Yanagisawa 1994). Though players who have had
a cleft palate repair early in life sometimes report problems with seal following periods
of increasingly intensive playing, there is no reason to assume that they are more prone
to this type of problem than other wind players.

Pharyngoceles and Laryngoceles

The pharynx (the throat) is a muscular tube whose walls, unlike those of the trachea
or the thorax, are not supported by a skeleton of cartilage or bone. The high pressures
required for playing double-reed and small-bore brass instruments can cause the phar-
ynx to become chronically distended (Levine 1986). Young players are likely to be
most at risk because its muscular walls are less well developed at this stage than in
adults. Consequently, it may be advisable to limit the volume of sound and amount of
playing that young musicians carry out in parts of the range where the pressures are
highest (Stasney, Beaver, and Rodriguez 2003). For brass instruments, this will be the
extreme upper end of the range, but the same is not true for all woodwind instruments
(Fuks and Sundberg 1999). Certain conditions in which the supporting connective
tissue of the pharynx is weak may increase the risk of distension, which can lead to
the formation within its walls of pocketlike folds called pharyngoceles (fig. 6-4D).
When these develop, swallowing can become difficult and food can be trapped within
them. As a result, food fragments may appear again in the mouth during playing when
the throat expands, even though some time has elapsed since eating. A small amount
of distension of the neck during playing is generally regarded as normal and often
makes it necessary to keep the top shirt button undone or to wear shirts with a larger
206 Chapter Six

Figure 6-4. A–C. Rupture of the orbicularis oris muscle, commonly known as Satchmo’s syn-
drome. The tear in the muscle (the purse-string muscle of the mouth) is typically at right angles
to the direction of its fibers. The affected part of the muscle, indicated by the box in A, is seen
at higher magnification in B. Underlying fatty tissue may push out through the tear. Nowadays
this can be treated surgically by stitching the tear closed (C). D. This diagram shows symptoms
of several problems that may arise as a result of the high pressure generated in the airway during
playing. Air may be forced into the parotid salivary gland, causing it to swell (pneumoparotitis).
The muscular wall of the pharynx may become distended (dashed line), forming a pharyngocele.
A small sac that is in communication with the central space of the larynx may become inflated so
that it appears on the outside of the thyroid cartilage. This is called a laryngocele.

collar size. Where distension is extreme, a medical opinion should be sought. In for-
mer times, some players used to strap the neck during playing, and even today a soft
cervical collar may be recommended (Smith and Levine 1986). An intensive program
to strengthen the superficial muscles of the neck (trapezius, sternocleidomastoid,
and platysma) is also likely to be suggested even though none of these lie directly in
contact with the pharynx. If there is a persistent and significant problem, switching to
another instrument is advisable.
Within the larynx there are also spaces that can become enlarged by pressurized air
(Levine 1986; Gallivan and Gallivan 2002). The gap between the vocal folds and the
vestibular folds is known as the ventricle. This is a slitlike opening that enlarges into
a blind-ended pouch the extends forward and is a weak point in the wall of the larynx,
which is otherwise supported by cartilage. It can become enlarged to form what is
known as a laryngocele, a cavity that pushes outward to emerge above the thyroid
cartilage (fig. 6-4D). The first clinical description of laryngoceles is thought to be
from Napoleon’s surgeon, Baron Larrey, who described them in army drill manuals
in 1829 (Gallivan and Gallivan 2002), but they appear to be very common in wind
players, in whom they are generally asymptomatic. They are an occupational hazard
for glassblowers, who also generate high airway pressures. Problems only arise when
the fluid that collects in them becomes infected.

Pneumoparotitis (Wind Parotitis)

Three sets of salivary glands empty into the mouth. The largest is the parotid gland,
which lies on the cheek over the angle of the jaw (fig. 6-4D). The saliva it produces
 The Embouchure and Wind Playing 207

runs along a duct that opens through the muscular wall of the cheek at the level of the
second upper molar tooth. Though large, the gland is not usually prominent unless it
becomes swollen, for example, during an attack of mumps. The high pressures in the
mouth during the playing of brass and woodwind instruments can drive air into the
duct, causing the gland to inflate. This has been recorded in wind players for hundreds
of years, beginning with medieval cornet players. French Foreign Legionnaires who
wanted to avoid duty were known to induce this condition so they appeared to have
the mumps (Levine 1986). High air pressure in the mouth can drive saliva as well
as air into the duct of the parotid gland. This can carry bacteria that may precipitate
infection and a blockage of the duct, leading to swelling of the gland. The swelling
is painful because the gland is enclosed in a tough sheath. Such infections are treated
with antibiotics but are not easily prevented (Yeo et al. 2002)

Syncope
Syncope is a lack of blood flow to the brain that may cause dizziness and fainting.
It can be caused by the very high pressures that are produced in the thoracic cavity
during the playing of brass instruments such as the trumpet. When players generate
pressures that exceed the peak (systolic) blood pressure, the vessels in the thorax be-
come compressed and the flow of blood to the brain is reduced or temporarily halted.
Syncope is most likely to occur in players of small-bore brass instruments, at the end
of long passages of high-intensity playing in the upper registers (Tucker, Faulkner,
and Horvath 1971).

Eye Problems
Some accounts of brass playing technique suggest that generating high pressure in the
airway may cause bleeding within the eye. This is not likely unless the player has an
existing eye problem such as diabetic retinopathy or age-related macular degeneration
(wet type), which are associated with the growth of new blood vessels that are fragile
and therefore at risk of rupture (Marmor 1986). Raised pressure in the blood vessels of
the head that occurs when playing some instruments has been demonstrated to lead to
a transient increase in the pressure within the eyeball together with an increase in the
thickness of a layer beneath the retina (choroid) that contains most of the blood vessels
(Schuman et al. 2000). Both of these increases last only for as long as high lung pres-
sure is maintained and fall immediately when the player takes a breath. Chronic raised
intraocular pressure is symptomatic of glaucoma, a condition in which the production
of the fluid within the eyeball exceeds its ability to drain away into the venous system
of the eyeball. It can ultimately lead to blindness due to compression of the head of the
optic nerve; however, for this to happen the high pressure must be maintained for pro-
longed periods of time (many days). As the paper by Schuman appears to be the only
rigorous study of its type performed to date, it is worth describing in more detail. Only
three subjects were involved. One was a seventy-two-year-old oboist and the second
a seventy-eight-year-old trumpeter, both of whom had been playing for fifty years or
more, while the third was a forty-six-year-old player of various wind instruments who
208 Chapter Six

had thirty-six years’ playing experience. Each had a different type of glaucoma though
this had little or no effect on their visual acuity, and their intraocular pressures when
not playing were close to normal (10–20 mmHg). In the oboist, intraocular pressure
rose from 16 to 40 mmHg over a period of twenty seconds when playing high and
loud. For another of the subjects, playing the saxophone or clarinet as loud and as high
as possible raised intraocular pressure only from 10 to 21 mmHg, though for neither
instrument is this the range where peak pressures are generated (Fuks and Sundberg
1999). When playing the trumpet, intraocular pressure in the same individual rose to
over 30 mmHg. Other studies have provided no evidence that intraocular pressure is
more sensitive to raised thoracic pressure in individuals with existing glaucoma than
in those without, so these values are probably typical for such instruments. In order
to increase the scope of the study, the authors carried out detailed visual field tests in
an additional forty-six professional musicians but did not measure their intraocular
pressure during playing. These were nine players of high-pressure wind instruments,
twelve of low-pressure instruments, and twenty-three players of nonwind instruments.
These players had on average more than thirty years’ playing experience. The high-
pressure wind players had an extremely small but statistically significant increase in
what could be regarded as glaucomatous damage, but at a level that was unlikely to
be clinically significant. Clearly this subject would benefit from investigation of a
much larger pool of players in which various aspects of lifestyle were also taken into
consideration.

Rupture of Orbicularis Oris (Satchmo’s Syndrome)

The orbicularis oris muscle can be physically damaged in brass instrumentalists
through the demands of playing in very high registers or with great intensity. Rupture
of the muscle is commonly known as Satchmo’s syndrome because just such an in-
jury was sustained by Louis Armstrong in 1935, with the result that he had to cease
playing for a considerable period. The injury is often triggered by a player switching
to a new mouthpiece requiring a modification of the embouchure (Papsin, Maaske,
and McGrail 1996). The most commonly reported symptoms are weakness of the
embouchure, followed by pain, loss of tone, and loss of seal against the mouthpiece.
The damage to the muscle is typically a tear that lies at right angles to the direction
of the muscle fibers (fig. 6-4A, B). If left untreated, this may heal only incompletely,
leaving a rigid scar, or may result in a thinning of the muscle at the site of injury. If
the tear remains open, the underlying fatty tissue may push through the gap as a small
hernia. Before the advent of surgical treatments, the only remedy was either to stop
playing and hope that it would heal naturally, to avoid extreme registers, or to try
to continue playing using a small-bore mouthpiece. Recovery was rarely complete.
Nowadays, simple surgical methods can be used to repair the damage, apparently with
a reasonable degree of success (fig. 6-4C). The operation is generally carried out with
the patient awake, as the incisions are superficial. Under such circumstances, asking
the patient to buzz the embouchure during the procedure makes it easier to identify
the damaged area. In addition, when the muscle is stitched but before the wound in
the skin is closed, the surgeon can check that the sutures will hold when the muscle
 The Embouchure and Wind Playing 209

is contracted to form the embouchure. Playing can generally recommence three to

twelve weeks after the operation, and players are said to return to their full level of
proficiency (Papsin, Maaske, and McGrail 1996; Planas 1988).

Focal Dystonia of the Embouchure

Focal dystonia is a loss of muscle control that can affect either the embouchure or the
hand in musicians. A full account of its nature and origin will be found in chapter 8.
The overall incidence of dystonia among musicians as a whole is difficult to quantify;
however, in a group of over 1,300 players with neuromuscular or musculoskeletal
problems that affected their playing, about 8 percent were given a diagnosis of dysto-
nia (Lederman 2003). The most detailed account to date of embouchure dystonia is a
review of twenty-six cases among brass and woodwind players (Frucht et al. 2001).
Because of the number of cases covered in this study, it is possible to draw some
preliminary conclusions concerning the incidence of different forms of embouchure
dystonia in players of a range of instruments, as well as to assess the outcomes of vari-
ous therapies. The dystonia manifests itself as a lack of embouchure control, which
may include symptoms such as lip tremor or fatigue, or involuntary activity of facial
muscles contributing to the embouchure or of those moving the jaw. In individuals
with lip tremor, this symptom appears with the onset of playing and is usually asso-
ciated not with the entire range of the instrument, but with a particular register. The
abnormal muscle activity often involves a lateral pulling on one corner of the lip.
Among the musicians studied by Frucht, this was reported only in players of high-
register brass instruments, most commonly the French horn. In players of low-register
brass instruments, involuntary lip closure was a more typical symptom. Like focal
dystonia elsewhere, the typical age of onset for embouchure dystonia is the mid- to
late thirties, on average around twenty-five years after the musician first started to
study the instrument. It is rarely associated with pain, though some discomfort is not
unusual. A variety of treatments have been tried for embouchure dystonia, such as
retraining, the use of dental prosthetic devices, and various alternative therapies, but
none has a satisfactory success rate. Botox injections are often advocated for reducing
the strength of the troublesome muscle contractions. These are rarely effective, as not
only do they not treat the root of the problem, but it is extremely difficult to deliver the
precise quantity needed to reduce the level of contraction to the correct degree in such
small muscles. As with other forms of dystonia, the symptoms are typically context
dependent (i.e., they are associated only with playing) and only rarely do they spread
to other activities involving the mouth. This is exemplified by a case quoted by Frucht
in which a player who gave up the trumpet because of dystonia successfully switched
to playing the bassoon without reactivating the symptoms.

Damage to Nerves in the Face

The muscles that contribute to the embouchure are supplied by branches of the facial
nerve. These emerge from under the parotid salivary gland, whose position is shown
in figure 6-4. The fine branches of the nerve are vulnerable to deep cuts to the face.
210 Chapter Six

As the nerves run predominantly horizontally, vertical cuts are likely to be the most
dangerous. If nerve branches are severed, the muscles will be paralyzed. If the ends
of the nerves can be brought together, the motor neurons supplying the muscles may
be able to regrow to supply the muscles. Alternatively, nerve grafts can be used to aid
regrowth. Paralysis of the facial nerve also occurs in a condition known as facial nerve
(or Bell’s) palsy. This condition is usually of unknown origin but may arise following
exposure to cold or be associated with dental treatment (perhaps as a result of injec-
tion of anesthetic) or a middle ear infection. The outcome depends on how severely
the nerves are damaged. Complete recovery may occur over a period of months, par-
ticularly if only the ends of the nerve branches have lost function. However, recovery
sometimes brings changes to the pattern of nerve supply so that attempts to activate
one muscle or set of muscles cause abnormal twitches in others.
Sensation to the face is supplied by the branches of the trigeminal nerve. Trigeminal
neuralgia, another condition that is usually of unknown cause, results in severe stab-
bing pain in the face. It is often triggered by gentle touch, by chewing, or by brushing
the teeth. Though each bout of pain lasts only a short time, it can nonetheless be de-
bilitating. The symptoms, although not the underlying cause, are usually controllable
with drugs. More drastic treatments involving surgical intervention to cut nerve fibers
have varying levels of success and of course will result in loss of sensation, which is
crucial for wind players. It has recently been discovered that compression of the base
of the trigeminal nerves by blood vessels within the skull can be a source of trigeminal
neuralgia, in which case it can be relieved by invasive surgical treatment. Burning
mouth syndrome (a painful sensation of heat within the mouth) may be related to tri-
geminal neuralgia in the sense that it also affects this nerve; however, if anything it is
even more enigmatic (Wall, McMahon, and Koltzenburg 2006). It is most commonly
reported in postmenopausal women but is not confined to this group. Emotional stress
is often cited as a trigger, though there is a danger that this conclusion is reached only
as the last stage in a process of elimination and is therefore a tacit admission that an
organic cause cannot be found. Changes in the properties of the nerve endings within
the mouth have recently been demonstrated in one group of patients, but it is still not
known why this occurs (Yilmaz et al. 2007).

REFERENCES

Amir, N. 2004. Some insight into the acoustics of the didgeridu. Applied Acoustics 65:1181–96.
Barbenel, J. C., J. B. Davies, and P. Kenny. 1986. Science proves musical myths wrong. New
Scientist 110 (1502):29–32.
Barbenel, J. C., P. Kenny, and J. B. Davies. 1988. Mouthpiece forces produced while playing
the trumpet. J Biomech 21 (5):417–24.
Basmajian, J. V., and E. R. White. 1973. Neuromuscular control of trumpeters’ lip. Nature 241
(5384):70.
Benade, A. H. 1986. Interactions between the player’s windway and the air column of a musi-
cal instrument. Cleve Clin Q 53 (1):27–32.
Brown, A. F. D. 1976. A cinefluorographic pilot study of the throat while vibrato tones are
played on flute and oboe. J Int Double Reed Soc no. 4.
 The Embouchure and Wind Playing 211

Cossette, I., P. Sliwinski, and P. T. Macklem. 2000. Respiratory parameters during professional
flute playing. Respir Physiol 121 (1):33–44.
Devroop, K., and K. S. Chesky. 2002. Comparison of biomechanical forces generated during
trumpet performance in contrasting settings. Med Prob Perform Artists 17:149–54.
Dibbell, D. G., S. Ewanowski, and W. L. Carter. 1979. Successful correction of velopharyn-
geal stress incompetence in musicians playing wind instruments. Plast Reconstr Surg 64
(5):662–64.
Engelman, J. A. 1965. Measurement of perioral pressures during playing of musical wind in-
struments. Am J Orthod 51 (11):856–64.
Fine, L. 1986. Dental problems in the wind instrumentalist. Cleve Clin Q 53 (1):3–9.
Fritze, C., and J. Wolfe. 2005. How do clarinet players adjust the resonances of their vocal
tracts for different playing effects? J Acoust Soc Am 118 (5):3306–15.
Frucht, S. J., S. Fahn, P. E. Greene, C. O’Brien, M. Gelb, D. D. Truong, J. Welsh, S. Factor, and
B. Ford. 2001. The natural history of embouchure dystonia. Mov Disord 16 (5):899–906.
Fuks, L., and J. Sundberg. 1999. Blowing pressures in bassoon, clarinet, oboe and saxophone.
Acustica 85:267–77.
Gallivan, K. H., and G. J. Gallivan. 2002. Bilateral mixed laryngoceles: Simultaneous strobo-
videolaryngoscopy and external video examination. J Voice 16 (2):258–66.
Gilbert, T. B. 1998. Breathing difficulties in wind instrument players. Md Med J 47 (1):23–
27.
Gordon, N. A., D. Astrachan, and E. Yanagisawa. 1994. Videoendoscopic diagnosis and cor-
rection of velopharyngeal stress incompetence in a bassoonist. Ann Otol Rhinol Laryngol
103 (8 Pt 1):595–600.
Gualtieri, P. A. 1979. May Johnny or Janie play the clarinet? The Eastman Study: A report on
the orthodontic evaluations of college-level and professional musicians who play brass and
woodwind instruments. Am J Orthod 76 (3):260–76.
Hart, C. W., and J. A. Logemann. 1986. Tonsils and adenoids and the professional musician.
Med Prob Perform Artists 1:58–60.
Herman, E. 1974. Dental considerations in the playing of musical instruments. J Am Dent As-
soc 89 (3):611–19.
———. 1981. Influence of musical instruments on tooth positions. Am J Orthod 80 (2):145–55.
Heuser, F., and J. L. McNitt-Gray. 1991. EMG potentials prior to tone commencement in trum-
pet players. Med Prob Perform Artists 6 (4):51–56.
———. 1993. EMG patterns in embouchure muscles of trumpet players with asymmetrical
mouthpiece placement. Med Prob Perform Artists 8:96–102.
———. 1998. Enhancing and validating: Pedagogical practice. The use of electromyography
during trumpet instruction. Med Prob Perform Artists 13 (4):155–59.
King, A. I., J. Ashby, and C. Nelson. 1988. Laryngeal function in wind instrumentalists: The
woodwinds. J Voice 1:365–67.
———. 1989. Laryngeal function in wind instruments: The brass. 3 (1):65–67.
Klotz, D. A., J. Howard, A. S. Hengerer, and O. Slupchynskj. 2001. Lipoinjection aug-
mentation of the soft palate for velopharyngeal stress incompetence. Laryngoscope 111
(12):2157–61.
Lederman, R. J. 2003. Neuromuscular and musculoskeletal problems in instrumental musi-
cians. Muscle Nerve 27 (5):549–61.
Levine, H. L. 1986. Functional disorders of the upper airway associated with playing wind
instruments. Cleve Clin Q 53 (1):11–13.
Marmor, M. F. 1986. Vision and the musician. Med Prob Perform Artists 1:117–20.
Nixon, G. 1963. Dental problems in brass instrumentalists. Br Dent J 105:160–61.
212 Chapter Six

Papsin, B. C., L. A. Maaske, and J. S. McGrail. 1996. Orbicularis oris muscle injury in brass
players. Laryngoscope 106 (6):757–60.
Planas, J. 1988. Further experience with rupture of the orbicularis oris in trumpet players. Plast
Reconstr Surg 81 (6):975–81.
Ridgeon, J. 1986. The physiology of brass playing: Manton, UK: Brass Wind Educational
Supplies.
Rindisbacher, T., U. Hirschi, B. Ingervall, and A. Geering. 1990. Little influence on tooth posi-
tion from playing a wind instrument. Angle Orthod 60 (3):223–28.
Rydell, R., M. Karlsson, A. Milesson, and L. Schalen. 1996. Laryngeal activity during wind
instrument playing: Video endoscopic documentation. Logoped Phoniatr Vocol 21:43–48.
Schuman, J. S., E. C. Massicotte, S. Connolly, E. Hertzmark, B. Mukherji, and M. Z. Kunen.
2000. Increased intraocular pressure and visual field defects in high resistance wind instru-
ment players. Ophthalmology 107 (1):127–33.
Schwab, B., and A. Schultze-Florey. 2004. Velopharyngeal insufficiency in woodwind and
brass players. Med Prob Perform Artists 19:21–25.
Smith, R., and H. Levine. 1986. Hypopharyngeal dilatation in musicians. Med Prob Perform
Artists 1:20–23.
Stasney, C. R., M. E. Beaver, and M. Rodriguez. 2003. Hypopharyngeal pressure in brass musi-
cians. Med Prob Perform Artists 14:153–55.
Tucker, A., M. E. Faulkner, and S. M. Horvath. 1971. Electrocardiography and lung function
in brass instrument players. Arch Environ Health 23 (5):327–34.
Wall, Patrick D., S. B. McMahon, and Martin Koltzenburg. 2006. Wall and Melzack’s textbook
of pain. 5th ed. Philadelphia: Elsevier/Churchill Livingstone.
Weikert, M., and J. Schlomicher-Thier. 1999. Laryngeal movements in saxophone play-
ing: Video-endoscopic investigations with saxophone players. A pilot study. J Voice 13
(2):265–73.
White, E. R., and J. V. Basmajian. 1973. Electromyography of lip muscles and their role in
trumpet playing. J Appl Physiol 35 (6):892–97.
Yeo, D. K., T. P. Pham, J. Baker, and S. A. Porters. 2002. Specific orofacial problems experi-
enced by musicians. Aust Dent J 47 (1):2–11.
Yilmaz, Z., T. Renton, Y. Yiangou, J. Zakrzewska, I. P. Chessell, C. Bountra, and P. Anand.
2007. Burning mouth syndrome as a trigeminal small fiber neuropathy: Increased heat and
capsaicin receptor TRPV1 in nerve fibers correlates with pain score. J Clin Neurosci 14
(9):864–71.
Ziporyn, T. 1984. Pianist’s cramp to stage fright: The medical side of music-making. JAMA
252 (8):985–89.
 Chapter Seven

The Structure and

Organization of the Brain

The human brain manages to compress a phenomenal amount of computing power

into a very small volume. Its main task is through our senses, to analyze the world
around us and as a result to generate appropriate behavior. In so doing it routinely
carries out computations in real time that the most powerful computers have dif-
ficulty in matching. On the other hand, before we start to feel too self-satisfied, it
must also be admitted that it has difficulties with simple arithmetic operations that
the cheapest of calculators can manage with ease. Nevertheless, complex sensory
tasks such as our ability to identify objects regardless of the angle of view, to recog-
nize individual faces or voices instantly or a piece of music from a phrase lasting a
couple of seconds, are remarkable feats that we perform effortlessly. Not only do we
take these complex tasks for granted, but the ability to lead a normal life is entirely
dependent on their being achieved with almost total reliability. Any lapses would
immediately be detected by our companions, who would quickly alter their behavior
toward us as a result. The brain’s achievements in controlling the movement of the
body are equally impressive. Think of a cricketer diving at full stretch to catch a ball.
The three-dimensional trajectory of the ball must first be calculated in a fraction of a
second and its subsequent motion predicted. The body is then moved toward a point
where the ball can be intercepted. If time is short the player may throw his body down
along its own predicted trajectory with an arm extended to reach the intercept point
with millisecond accuracy. While all this is happening the brain is also occupied with
maintaining normal body functions (breathing, temperature regulation, metabolic
control, etc.), providing emotional responses to the situation, matching the experi-
ence with previous memories, and predicting how success or failure with the catch
will affect the outcome of the match!
The brain weighs approximately 1.5 kilograms. Forty percent of this is made up
of the nerve cells involved in carrying out computations or memory storage, while
60 percent represents the cabling that links its different regions together. It contains
approximately 100 billion (1011) nerve cells (or neurons), each of which is connected
to hundreds or thousands of other nerve cells. The structure of typical nerve cells in
the brain is shown in figure. 7-1. Each has a cell body at the center of which lies the

213
214 Chapter Seven

Figure 7-1. The main features of nerve cells in the brain. Each nerve cell has a cell body that
contains the nucleus, a spherical structure that carries all the information the cell requires to
maintain its normal functions. The cell body receives information from other nerve cells, either
via contacts (synapses) made directly onto it or via its many branching dendrites. The cell body
also has a single long axon that carries information to other nerve cells. Axons are often sheathed
in myelin, which is composed of spirally wrapped glial cell membranes, and this speeds the pas-
sage of information along them. Axons end in synapses, which contact other neurons. Synapses
contain many small vesicles that carry the chemical neurotransmitters that the nerve cells re-
lease in order to communicate with each other. They also contain mitochondria, which generate
the energy needed to operate the synapse. In addition to projection interneurons, which carry
information over long distances, the nervous system contains very large numbers of small local
interneurons, which are responsible for the information processing that takes place within the
computational units of the brain.

nucleus, the repository of the genetic instructions that are necessary for the develop-
ment and normal functioning of the cell. These instructions carry the blueprints for the
manufacture not only of the proteins that contribute to the structure of the nerve cells,
but also those that control the chemical reactions and signaling processes that take
place between and within the nerve cells (enzymes, receptors, and signal molecules).
The proteins are manufactured almost entirely within the cell body and so when
needed elsewhere, they must be transported, sometimes over considerable distances,
to these sites. Many branches emerge from the cell body. Most of these are dendrites,
whose role is to receive information through contacts (synapses) with other nerve
 The Structure and Organization of the Brain 215

cells in the nervous system. Each nerve cell also has a single branch, the axon, which
carries information to other nerve cells. In some neurons it can be a meter or more in
length, though in others it may terminate quite close to the cell body. The axon ends
in a series of terminal branches carrying the synapses that contact other nerve cells in
the circuit (fig. 7-1).
The nerve cells of the brain can be classified into four main categories. Sensory neu-
rons (fig. 1-3) carry information from the skin, muscles, and joints and are responsible
for sensations such as touch, pain, and temperature. Sensory cells, albeit of a rather
different form, are also found within the special sense organs and are responsible
for the sensations of taste, smell, vision, hearing, and balance. We interact with the
world through our verbal and physical responses to it, and our behavior is ultimately
manifested by the action of muscles, which are controlled by motor neurons (fig. 1-3).
Between the sensory neurons and the motor neurons lie the great computational centers
of the brain. Within these, the analysis and interpretation of the information received by
the nervous system are carried out by what are known as interneurons—neurons that
link one nerve cell to another (figs. 1-3, 7-1). There are two classes of interneurons.
Those in one group (projection interneurons) have long axons that carry information
over considerable distances from one part of the brain to another. The second group
contains local interneurons, which are much smaller and confined to small, well-defined
regions of the brain devoted to a particular task. The local interneurons are the ones that
carry out the complex computations of the brain; the more advanced the animal, the
greater the proportion of its total population of nerve cells that are of this type.
Each nerve cell is linked to hundreds or thousands of others, and the connections
between pairs of nerve cells typically comprise tens or hundreds of individual syn-
aptic connections. It has been estimated that a typical neuron receives 1,000–10,000
synapses, though for some neurons the figure may be as high as 200,000. The total
number of synapses in the cortex of the brain alone is conservatively estimated at 60
trillion (6x1015)! The nerve cells communicate with each other by releasing chemicals
called neurotransmitters, which are stored in vesicles that are found at the synapse
(fig. 7-1). The neurotransmitters may excite or inhibit a neuron, or alter its sensitiv-
ity to other inputs. There are many tens of different types of neurotransmitters and
each can have several quite distinct effects on nerve cell activity. All of this serves to
increase the subtlety of the interactions between neurons. In addition, the connections
do not remain immutable throughout life but can be broken and re-formed over time
to change the properties of brain circuitry so that it can adapt to different situations.
From this battery of statistics it will be obvious that the brain is literally unimaginably
complex; one might be forgiven for wondering if there is any hope at all for attempts
to unravel how it operates. We have in fact made quite a good beginning in the last
century and a half toward revealing its secrets. In the last fifty years progress has
become increasingly rapid through the development of a constantly expanding range
of new investigative techniques that modern science has made available. However, it
would be naive to imagine that we have done more than make a respectable start to
our exploration of the brain. Today, the brain is studied at many different levels. It is
possible to examine individual connections between nerve cells in order to reveal their
properties or to look at how small groups of neurons interact with each other. This
216 Chapter Seven

information may be useful in the search for drugs to treat illness or to gain an insight
into how information is coded within the nervous system, but it does not often reveal a
great deal about how the brain achieves what it does in analytical or behavioral terms.
At the other end of the spectrum, it is possible to see which areas of the brain are ac-
tive during particular types of behavior or when we are carrying out certain mental
tasks; however, it has not been revealed how this is achieved at a cellular level or how
the outcome is incorporated into normal behavior. Finally, one can also simply treat
the brain as a black box and study our psychological responses to different situations,
but once again this will not reveal the underlying mechanisms. Because of the incred-
ibly complex nature of the interactions between nerve cells, bridging the gap between
these different approaches raises enormous challenges, and in any attempt to describe
how the brain responds in a particular context (such as music), these gaps in the fabric
of our understanding soon make their presence felt only too clearly.
Strange as it may seem, neurons are not the most abundant type of cell in the ner-
vous system. They are outnumbered ten to fifty times by another class of cell called
glia. There are several types of glial cells, whose basic role is to support the function
of the nerve cells—to make the brain a “land fit for neurons.” One class of glial cell
(astrocytes) regulates the environment of the neurons. Within the brain, the small
spaces between the nerve cells are filled with a salty fluid through which nutrients,
oxygen, neurotransmitters, ions, and other chemicals required for nerve cell survival
can diffuse freely. The glial cells control the composition of the salts in this fluid to
optimize it for nerve cell performance. They also break down some neurotransmitters
that would become toxic if they were not quickly removed after they had performed
their function. They help to provide general structural support for brain tissue and
maintain the integrity of key elements such as the synapses. Astrocytes also play a
role in regulating blood flow so that this is matched to the level of activity in regions
of brain tissue on a second-by-second basis.
Another class of glial cell (the oligodendrocytes or Schwann cells) produces sheets
of membrane that wrap around the long axons that link the nerve cells together. This
covering, known as the myelin sheath, acts as an insulator and has the effect of in-
creasing the speed at which signals can pass along the axon (fig. 7-1). It is found only
in vertebrates and is particularly important in large animals. One reason it is so hard
for us to catch or swat a fly with our hand is that because the fly is so small, the axons
of its nerve cells are much shorter than ours. Even with the aid of myelination, the
time taken for signals to travel from the brain to the muscles used to move the hand is
much longer than the time it takes for a signal to travel from the brain of the fly to the
muscles that move its legs and wings, and we can only hope to outwit it using guile.
Bundles of myelinated axons appear white in brain tissue and as a result are referred to
as white matter (fig. 7-4). This therefore represents the cabling of the nervous system.
The areas of the brain that contain groups of nerve cell bodies and dendrites are darker
and as a result are referred to as gray matter (fig. 7-4). The gray matter areas are the
computational centers of the brain and it is here that the incoming streams of informa-
tion are passed via synapses to the local interneurons, which integrate and manipulate
the signals before forwarding the data for further analysis elsewhere. Gray matter
areas may appear as solid masses or as thin sheets, depending on their location.
 The Structure and Organization of the Brain 217

A third class of glial cell (microglia) is involved in fighting disease and removing
the remnants of brain cells that have become damaged or have died. Very few nerve
cells are produced beyond the first few months after birth, but glial cells continue to
be replaced throughout life. As a consequence, brain tumors (which form as a result
of uncontrolled cell division) never develop from nerve cells but arise either from glial
cells or from the membranes that enclose the brain (the meninges).
Though the brain represents only about 2 percent of total body weight, it is a very
active tissue; it uses 15–20 percent of the oxygen supply to the body and receives a
similar proportion of the total blood supply. This varies little, whether we are trying
to solve a difficult mathematical or chess problem or learn a complex new piece of
music or are simply watching television or sleeping in our favorite armchair. There
are many reasons for this. First, a considerable amount of brain activity is concerned
with housekeeping functions. We can regard the body rather like an astronaut’s
spacesuit, which is responsible for life support; much of the brain is devoted to regu-
lating such activities. These include the control of breathing, heart rate, temperature
regulation, water balance, digestion, excretion, metabolism, and so forth. Second, as
we have already seen, many perceptual and motor control functions that we take for
granted require considerable brain activity, and these are constantly ongoing. Even
when we are sleeping, the brain is busy and not only in dreaming. The role of sleep
is still only partly understood, but we all know the consequences of going without it.
These include loss of perceptual skills, decreased reaction times, and ultimately even
disorientation and hallucination coupled with an increasing tendency to lapse into
unconsciousness as sleep finally becomes irresistible.
The substance that makes up the living brain is very soft, with a consistency simi-
lar to that of blancmange. This potentially creates something of a problem because
if unsupported, the lower surface of the brain would be compressed against the skull
by its own weight. This would be sufficient to collapse blood vessels, cutting off the
supply of oxygen and nutrients, and death of the tissue would rapidly ensue. The rea-
son this does not happen is that the brain floats in a salty liquid called cerebrospinal
fluid (CSF). As a consequence, its effective weight is reduced from 1.5 kilograms to
about 50 grams. The CSF also resists the tendency of the brain to move within the
skull and make contact with the bone. This is very effective unless we receive such a
forceful blow to the head that buffering by CSF cannot prevent the brain from striking
the skull. Such an impact may result in a partial or complete loss of consciousness,
bleeding, or other damage that can cause the brain to swell and to become compressed.
The brain and spinal cord are also protected by being wrapped in three layers of
membranes known as meninges. The inner layer (the pia) is stuck to the surface of the
brain and gives it some support. The CSF lies between this layer and a second layer
(the arachnoid) that is very fine but impervious to fluid. This layer lies in contact with
a very tough outer layer (the dura) that has the consistency of parchment and is fused
to the inside of the skull. All three layers extend down through the central cavity in
the vertebral column that contains the spinal cord. If it is suspected that the meninges
have become infected (meningitis), CSF may be sampled by collecting it from the
space within the vertebrae of the back (lumbar puncture, or spinal tap) so that it can
be examined to determine the nature of the infection or damage.
218 Chapter Seven

The object of this chapter is to provide an overview of the structure of the brain.
It will focus particularly on the structures that are important for an understanding of
the ideas presented in the subsequent chapters on the brain’s adaptation to music, the
auditory system, and the role of the nervous system in stress responses. As a conse-
quence, it should not be regarded as a general summary of brain organization, as many
important and interesting features that are not directly relevant to these ends will be
omitted.

Figure 7-2
 The Structure and Organization of the Brain 219

The brain originates in the embryo as a hollow tube, and in the mature brain the core
of this tube is represented as a series of cavities called ventricles (fig. 7-2C). These are
filled with cerebrospinal fluid, which is produced in the ventricles but which is able
to find its way out of the brain into the spaces between the layers of the meninges.
When we view it from the side, our impression of the intact brain is dominated by
the two cerebral hemispheres (fig. 7-2A), a pair of highly infolded masses of nervous
tissue that sit side by side, filling the upper part of the skull. At the back, beneath the
cerebral hemispheres, sits a single unpaired structure called the cerebellum (literally,
the little brain), which also has a highly folded surface. This is attached to the stalk,
known as the brain stem, that connects the hemispheres to the spinal cord. For the
purposes of anatomical description, the brain is usually divided into three parts: the
forebrain (which includes the cerebral hemispheres), the midbrain, and the hindbrain
(fig.7-2C). The latter two divisions together constitute the brain stem. The hindbrain
is made up of three components, the medulla oblongata (which is continuous with the
spinal cord), the pons, and the cerebellum. The pons is broader than the rest of the
hindbrain because it contains several massive bundles of axons that connect it to the
cerebellum. Traditionally, the cerebellum has been thought to be almost exclusively
concerned with the control of movement; however, this idea is beginning to change
as evidence accumulates for its involvement in a much broader range of functions,
including the analysis of musical sound. The cerebellum receives information from
three main sources: 1) the vestibular system, which is the part of the inner ear con-
cerned with balance, 2) the specialized sensory structures within muscles and tendons
that provide information on muscle contraction and loading, and 3) regions of the
cortex involved in muscle control. The core of the cerebellum is composed mainly of
the bundles of axons that connect it to the rest of the brain (i.e., white matter). The
computational part is a thin sheet of gray matter that is tightly folded to pack as much
of it as possible into the space available. Nerve cells of different types lie in a highly
organized array within this sheet. The basic principles of its design have been known
for decades; however, working out how it is used in the maintenance of equilibrium

Figure 7-2. The general architecture of the brain. A. The cortex is divided into regions called
lobes, which are named after the overlying bones of the skull. Its surface is made up of a com-
plex but relatively standardized series of ridges (gyri) separated by deep grooves (sulci). One of
these grooves, the central sulcus (broken line), is an important landmark, forming the boundary
between the frontal lobe and the parietal lobe. The occipital lobe lies at the back of the brain, and
the temporal lobe runs along the side, partly beneath the lower edges of the frontal and parietal
lobes. Underneath the cortex lies the brain stem. The parts of the brain stem visible here are the
cerebellum, which is connected to the pons, and the medulla. Together these three structures
form the hindbrain. B. The cortex as seen from above. Its two halves are called the cerebral
hemispheres and they are linked by a very thick bundle of nerve fibers known as the corpus cal-
losum. C. Here the brain has been cut in two by sectioning the corpus callosum and dividing the
brain stem along the midline, revealing a series of fluid-filled spaces called ventricles that lie at
its core. It is now possible to see the midbrain, which sits in front of the hindbrain. The forebrain
is composed of the cortex and a number of structures that lie just beneath it (see fig. 7-4). Two
of these (the thalamus and the hypothalamus) lie just beneath the wall of one of the ventricles.
The hypothalamus controls the pituitary gland, a structure that influences many important bodily
functions by releasing hormones into the bloodstream.
220 Chapter Seven

Figure 7-3. The brain stem contains the control centers for the life support systems of the body.
It exerts much of its influence via a series of ten cranial nerves. Two other cranial nerves (the
olfactory and the optic) arise from the forebrain. This diagram shows the position of the other
cranial nerves that emerge from the brain stem. Their names and a summary of their major func-
tions are given in table 7-1. Each cranial nerve is assigned a Roman numeral that is often used as
a shorthand label.

and in motor learning has proved a difficult problem to resolve. In front of the cerebel-
lum and pons lies the midbrain, which in humans is almost completely hidden under
the back of the cortex. The midbrain roof is composed of two pairs of swellings called
colliculi (Latin for “little hills”), which are involved in visual and auditory reflexes.
These cause us to turn our head very quickly to look toward a novel source of sound
or toward something that has appeared in the periphery of our visual field, actions that
were very important for the survival of our vertebrate ancestors.
The brain stem contains many axon bundles that carry information between the
spinal cord and the cortex, but its major role is as a control center for many of the
body’s life support systems. Even if all activity within the cortex ceases (as happens
in some forms of brain death), the basic functions required for the body to survive
can be maintained if the brain stem remains intact and continues to operate. It does
this through signals distributed via a series of cranial nerves that emerge from each
 The Structure and Organization of the Brain 221

Table 7-1. Cranial Nerve Function

No. Name Major Roles Main Functions in Music
I Olfactory Smell
II Visual Sight Score reading

}
III Oculomotor Eye movement (all); focus,
IV Trochlear control of pupil (III)
VI Abducens
V Trigeminal Sensory to face, jaw movement, Lip and tongue sensation, jaw
tenses soft palate position, helps maintain seal of
palate
VII Facial Facial muscle control,taste, Embouchure control, moistens
salivary glands mouth
VIII Vestibulocochlear Hearing and balance Hearing
IX Glossopharyngeal Salivary glands, taste, monitors Moistens mouth, supplies one
blood composition throat muscle
X Vagus Innervation of gut, larynx, heart, Laryngeal sensation, control of
airway, pharynx, palate laryngeal muscles, control of
shape of vocal tract and (with
V) seal of palate
XI Accessory Vagus + controls some neck Affects head/neck posture and
muscles assists deep breathing
XII Hypoglossal Supplies tongue muscles Control of tongue

side. There are twelve cranial nerves in total, and all but two arise from the brain
stem. These are shown in figure 7-3 and their names and functions are summarized in
table 7-1. Three of them control the movement of the eyes, plus the focusing of the
lens and the opening and closing of the pupils in response to light. One large nerve
(the trigeminal nerve) carries sensation from the face and controls the muscles used in
chewing; another (the facial nerve) controls the muscles of facial expression, which
are also responsible for forming the embouchure. The auditory nerve (more correctly
known as the vestibulocochlear nerve) carries both auditory sensation and information
from the vestibular apparatus of the inner ear, which is important for maintaining bal-
ance. Several of the nerves contribute to the sensation of taste and the control of the
salivary glands, while another (the hypoglossal nerve) is dedicated to the control of
the muscles of the tongue. One cranial nerve with a particularly large sphere of influ-
ence is the vagus nerve. This controls the larynx and some of the muscles of the throat
(pharynx) and has an influence over how easily air can flow into the lungs through
the bronchioles. It contributes to the regulation of heart rate, controls the movement
of the esophagus, stomach, and much of the gut, and promotes digestion. The core
of the brain stem therefore contains centers involved in all of these functions and
more. It will be clear from this account that a number of cranial nerve functions are
crucial to musical performance (see table 7-1). Many tasks require the coordination of
activity in several cranial and other nerves through these brain stem centers. A prime
example of this is the apparently simple act of swallowing. This requires closure of
the mouth (trigeminal nerve), backward movement of the tongue (hypoglossal nerve)
and lifting of the soft palate (glossopharyngeal and trigeminal nerves), raising of the
222 Chapter Seven

larynx (spinal nerves in the neck) and closure of the vocal and vestibular folds (va-
gus nerve), and contraction of the pharyngeal muscles (vagus) to push food or liquid
down the esophagus. Swallowing must also be coordinated with breathing—we make
a short intake of breath just before it starts so that there is a positive pressure in the
lungs in case stray material has to be expelled from the airway. The act of swallowing
therefore requires the orchestrated activity of many different centers within the brain

Figure 7-4
 The Structure and Organization of the Brain 223

stem. Failure of this system can lead to material ending up in the lungs or trachea, with
possibly fatal consequences.
Sensory information from the cranial nerves and from the spinal cord is sent to the
forebrain. In addition, however, the forebrain receives information directly from two
other cranial nerves: the optic and olfactory nerves, which carry visual information
from the eye and the sense of smell from the inside of the nose. These nerves differ
fundamentally from the other cranial nerves in that they arise from two complex struc-
tures that are derived from the brain during embryonic development. These are the
retina, which lines the back of the eyeball, and the olfactory bulb. The olfactory bulb
sits on the floor of the skull just above a specialized area at the back of the nose that
contains nerve cells that respond to odor. The retina and olfactory bulb contain many
interneurons, which process the raw visual and olfactory information before sending
it to the brain.
The forebrain has two main divisions. The first lies in continuity with the brain stem
and contains two major structures, the thalamus and the hypothalamus (figs. 7-2C and
7-4A). These are both paired structures, one of each lying on either side of the slitlike
third ventricle, which is found at the midline of the forebrain. Both structures are com-
posed of many separate clusters (or nuclei) of nerve cells, each with its own particular
function. The thalamus has three main roles. First, it is the main gatekeeper controlling
the flow of sensory information into the cortex. When we are asleep, for example, the
thalamus can block the inflow of sensation. Second, the thalamus contains nuclei that
are part of the complex network of brain areas that control movement. Finally, another
set of nuclei plays a role in emotional responses to sensation, instinctive behavior, and
general levels of awareness (arousal).
The hypothalamus lies beneath the thalamus and is another element of the life sup-
port system of the brain. It monitors body temperature and substances carried by the
blood such as circulating hormones like androgens and estrogens. It controls cellular
and whole body metabolism in two main ways: 1) It regulates the release of hormones

Figure 7-4. These diagrams show some of the large structures that lie deep in the forebrain,
beneath the cortex. In A, the far side of the brain is shown intact, together with the thalamus
and hypothalamus of the near side. The thalamus has several distinct regions with very different
roles. Some control the flow of sensory information to the cortex, while others are involved in
the control of muscle activity or of general levels of awareness. The hypothalamus monitors vital
functions both by analyzing the properties of the blood that passes through it and by receiving
sensory information from elsewhere in the body. It exerts its influence on metabolism and ho-
meostasis through the nervous system (e.g., via some of the cranial nerves and the sympathetic
nervous system) and through the release of hormones from the pituitary gland. In B we see one
of a group of structures collectively known as the basal ganglia, which play a key role in motor
control. In C are the principal components of the limbic system (see chapter 10); the hippocam-
pus and amygdala sit near the tip of the temporal lobe, while the cingulate gyrus lies immediately
above the corpus callosum. In D and E the positions of the structures shown in A and B can be
seen in two horizontal sections through one cerebral hemisphere. The asterisks mark where the
corpus callosum has been cut in order to separate the two hemispheres. The planes of these sec-
tions are shown by the broken lines on the small image of the brain. The corpus callosum and
the other areas of white matter represent the cabling that links different parts of the brain. The
thin outer layer of the cortex and the thalamus and basal ganglia are examples of the gray matter
areas in which the brain carries out computations.
224 Chapter Seven

from the pituitary gland. This large gland hangs by a stalk from the underside of the
brain just beneath the hypothalamus. 2) It activates the nerves that make up what is
called the autonomic nervous system. We will consider this division of the nervous
system in more detail when we deal with stress in chapter 10. At present it suffices to
say that it regulates such important functions as heart rate, blood flow, and digestion.
Some of the cranial nerves belong partly or entirely to the autonomic nervous system,
but other autonomic nerves arise from the spinal cord. The control exercised by the
hypothalamus operates over several time scales. It adjusts body temperature on a min-
ute-by-minute basis through its influence on activities such as sweating and shivering
and by switching blood flow to or from the skin. On a longer time scale, it regulates
the mobilization of stored energy reserves and the rate at which cells burn energy-rich
molecules, and influences food and water intake through the sensations of hunger
and thirst. It is influenced by the daily fluctuation of light levels to adjust metabolic
rate so that this is greatest during the day and lowest at night. When we travel rapidly
across several time zones, we suffer the consequences of putting this system out of
kilter. We may experience high body temperature during the night, which makes sleep
difficult, and low body temperature during the day, with the result that we feel cold
and shivery. As we adjust to the new time zone, our body temperature (a reflection of
metabolic rate) becomes synchronized with the local cycle of light and dark. On a still
longer time scale, the hypothalamus regulates the monthly cycle of female hormone
release and controls the pattern of growth during childhood and adolescence through
its influence on the release of growth hormone from the pituitary gland.
The second division of the forebrain is composed of the cortex together with some
other large masses of nervous tissue that lie beneath it and which constitute a major
part of what is known as the basal ganglia. As will be seen in the next chapter, the
basal ganglia are involved predominantly in the control of movement and comprise a
network of nuclei that also includes masses in another part of the forebrain and in the
midbrain. The midbrain component is called the substantia nigra and is the region of
the brain that degenerates in patients with Parkinson’s disease.
The cortex is made up of two hemispheres, one on each side of the brain, that are
linked together by a very large bundle of axons called the corpus callosum (fig. 7-
2B), which will feature significantly in the next chapter. The cortex is divided into
four major areas or lobes that are named after the bones of the skull under which they
lie (fig. 7-2A). Lying just behind the forehead is the frontal lobe, and moving pro-
gressively farther back we encounter first the parietal lobe and finally the occipital
lobe. Projecting downward from the parietal lobe rather like the thumb of a boxing
glove is the temporal lobe. The surface of the cortex is thrown into a series of folds
called gyri, separated by grooves known as sulci. If we were to make a horizontal
cut through the cortex, we would see that it is composed of a thin sheet of gray mat-
ter sitting on a large mass of white matter (fig. 7-4). The sheet of gray matter has
an organized structure composed of six sandwichlike layers, each of which has a
different function and pattern of connections. The white matter wires these to other
areas of the cortex, as well as to the rest of the brain and the spinal cord. The amount
of white matter required for this is indicative of the extent and complexity of the
connections that the cortex makes. The cortex is folded into sulci and gyri in order
 The Structure and Organization of the Brain 225

Figure 7-5. The primary motor and sensory areas of the cortex are here shown from three dif-
ferent viewpoints. It is through these regions that information from the eyes (the primary visual
cortex) and the ears (primary auditory cortex) and sensation from the skin and muscles (somato-
sensory cortex) are distributed for higher-level processing. The somatosensory area lies in the
parietal lobe, immediately behind the central sulcus, while the primary motor cortex (frontal cor-
tex), which is the main center directing the conscious control of the muscles, lies just in front.

to pack as much of the thin sheet of gray matter as possible into the limited volume
afforded by the skull.
The pattern of sulci and gyri is quite consistent between brains, and some of them
mark important boundaries between different functional areas of the cortex. The most
significant in this regard is the central sulcus, which lies at the boundary between the
Figure 7-6
 The Structure and Organization of the Brain 227

Figure 7-6. The organization of sensory and motor maps in the cortex. A. This image represents
a slice through the primary sensory cortex along the line marked A in the diagram of the brain
(D). Though the map of the body is clearly recognizable, it is distorted because proportionately
more of the cortex is devoted to the areas of the body that contain the greatest density of sensory
nerve endings. This is shown more graphically in C, which is a three-dimensional view of what is
known as the sensory homunculus. The overall organization of the motor map is shown in B. In
both panels A and B, the cranial nerves that contribute to the maps are indicated by name and
number (see fig. 7-3 and table 7-1). Those who are familiar with drawings of the homunculi will
notice some differences between this figure and standard representations, which have remained
virtually unchanged since first presented by Penfield and Rasmussen (1950): 1) the size of the
tongue is increased (in accord with the original observations of Penfield and Boldrey (1937), 2)
the genitals are shown in somatotopic order and the foot does not lie on the medial face of the
cortex (this derives from recent observations [Kell et al. 2005]), 3) the distal joint of the finger has
a larger representation and the palm a smaller one, 4) in C, the representation of the dominant
hand (here the right) is larger (Amunts et al. 1996), and 5) the unaccountably large representation
of the ears on the standard version of this figure is reduced. It should be noted, however, that
while the sensory homunculus is a reasonable interpretation of the organization of the primary
sensory cortex, there are problems in depicting the primary motor cortex in this way (as Penfield
and Rasmussen pointed out), and these are discussed in the next chapter.

frontal and parietal cortices (fig. 7-5). The different regions of the cortex have differ-
ent functions, and the central sulcus marks the boundary between the areas involved in
muscle or motor control (which lie in the frontal cortex) and those processing sensory
information (which lie in the parietal cortex and beyond). The ridge or gyrus immedi-
ately in front of the central sulcus is called the primary motor cortex (figs. 7-5, 7-6).
This contains a more or less orderly map of the muscle groups of the body but on the
opposite side. Hence the right motor cortex controls the movement of the left side of
the body and vice versa. The region of the primary motor cortex that lies just above
the temporal lobe is associated with movement of the face. Above this lies a large
area controlling the hand followed by areas controlling the arm, the trunk, and the leg
(fig. 7-6). In front of the primary motor cortex are some other motor areas with more
complex functions, which will be discussed in the next chapter. These include Broca’s
area, which is involved in generation of the motor program required for speech, which
is then “played” through regions of the primary motor area responsible for the muscles
of the mouth, tongue, larynx, and pharynx.
The ridge or gyrus immediately behind the central sulcus contains the primary
sensory cortex. This has a sensory map of the surface of the body that sits in register
with the primary motor map that lies on the other side of the sulcus (fig. 7-6A). Thus,
the face is represented on the most lateral surface, with the hand represented just
above, followed by the arm, trunk, and leg. The primary sensory cortex is the gateway
through which the sensory information from the body first enters the cortex, and as
with the primary motor cortex, the map represents the opposite side. Other discrete
primary sensory areas receive visual and auditory input (fig. 7-5). The primary audi-
tory area lies on the upper surface of the temporal lobe, while the primary visual area
is found at the back of the brain in the occipital lobe. Each also contains a sensory
map. In the case of the auditory cortex, this is one of sound frequency, and for the
visual cortex, it is a map of the visual field.
228 Chapter Seven

Now that we have identified the main primary motor and sensory areas of the
cortex, it is clear that most of the cerebral hemispheres remain unaccounted for in
terms of function. The information reaching the primary sensory areas is presented in
a very basic form. We do not just look, hear, touch, or see; we interpret, analyze, and
integrate this information. Sound may represent an object, an animal, or a person. It
may be music and be analyzed as such or be treated as language and broken down and
interpreted according to grammatical rules. We also combine information from the
different senses, bringing together texture, form, color, smell, movement, and sound
to generate a holistic view of the world. We may recognize an individual person,
animal, thing, or place by any one or by several of these features when we encounter
them or when we recall them from memory. This all takes place in the areas between
the primary cortical regions for the different senses; for this reason they are described
as association areas. It is possible to assign a particular role to some of these areas.
For example, a region known as Wernicke’s area, which lies just behind the primary
auditory cortex, contributes to our understanding of language as well as our ability
to generate it. Though the speech of patients with damage to this region sounds flu-
ent and grammatical, its content is often meaningless. One must be careful not to be
too prescriptive in assigning function to cortical regions, however, as our knowledge
of what they do (often based on the symptoms of brain damage) may be incomplete.
Broca’s “speech motor” area, though located in a region known primarily to be associ-
ated with the generation of movement, is also involved in grammatical construction
of spoken and written language as well as in verbal fluency. In the last two decades it
has also been discovered that Broca’s area contains what are known as mirror neurons:
neurons that are active both when an action is being performed by the brain’s owner
or when it is seen or even heard being performed by others. This may reflect an im-
portant role in the learning by imitation not only of speech, but also of other cultural
or noncultural types of movement.
The way in which various aspects of sensation are analyzed by the brain can be
surprising. Though we have a clear holistic visual impression of, for example, a person
walking across a room, different aspects of the scene are analyzed in discrete areas
of the visual system before being recombined. This task specificity of computational
units is common in sensory systems and indeed is often also a feature of complex
computer systems. The higher processing of visual information is sometimes de-
scribed as following two streams within the cortex. The first flows upward from the
primary visual cortex into the parietal cortex and beyond into the motor regions. This
is concerned primarily with the spatial relationships between objects: the “where” of
vision, which plays a role in the visual guidance of movement. The second stream
flows down into the temporal lobe and is more concerned with the identity of objects:
the “what” of vision. For example, there are regions involved in identifying not only
individual faces, but also what constitutes a face, so strange as it may seem, damage
to this region may make it impossible to distinguish a face from a roundish physical
object such as a hat (Sacks 1998). The primary visual cortex is interested mainly in the
angles of the lines that make up the contours of shapes. Color is analyzed elsewhere,
and it is possible (though rare) for brain damage to lead to loss of this aspect of vision
alone. A description of the surreal consequences this has is given in a fascinating book
 The Structure and Organization of the Brain 229

by the neurologist Oliver Sacks (Sacks 1995). Interestingly, not only is the ability to
see in color lost, but the ability to remember in color (i.e., to extract color information
from replayed memories) disappears as well. To take another example, subjectively
we would expect that some part of the visual system would be interested in the form
of objects, animate or otherwise, and in this we would be correct. However, it is not
so intuitive to suspect that the form of moving objects would be treated as a separate
category for analysis, which is actually the case (Zeki 1992). Damage to the region
of the cortex responsible for this function leads to a bizarre inability to see objects
or people when they are moving but not when they are stationary (Sacks 1995). This
raises an important question when we begin to consider how the brain responds to
music: does the brain always use the same strategies that we apply subjectively to
the analysis of music, or does it sometimes employ a different approach? If we fail to
answer this correctly and design our experiments accordingly, our investigations may
yield conflicting or ambiguous results. For many elements of musical interpretation
this question remains unresolved.
We now come to a system within the brain that is ancient and which is involved
in controlling drive-related behaviors that are crucial for survival, such as maternal
behavior and self-defense, but which as a result is also the seat of fear and anxiety
(see chapter 10). This is the limbic system. It comprises a number of areas of special-
ized cortex with links to the thalamus and hypothalamus, through which it initiates
autonomic responses to stress and fear. Its main cortical areas are the cingulate gyrus,
a band of gray matter that lies immediately above the corpus callosum, and the hip-
pocampus and amygdala, which lie near the tip of the temporal lobe (fig. 7-4C), in
the floor and at the end of the lateral ventricle, respectively. The latter two regions are
concerned with memory. The hippocampus is responsible for the laying down of as-
sociative memory (factual learning), though it is not the repository of memories once
they have been established. The amygdala, on the other hand, gives memories their
emotional flavor. Merely thinking about a stressful event can start the heart racing and
make us break out in a cold sweat; that is the responsibility of the amygdala. As we
will see in chapter 9, the amygdala is also involved in emotional responses to music.
It may come as a surprise for the reader to learn that though the two hemispheres ap-
pear generally similar in form, they are not identical in function. Most people consider
themselves right or left handed; that is, they have greater control over one hand or the
other and consequently use it for the tasks that are most demanding, such as writing.
This difference between the two sides of the brain arises during embryonic develop-
ment and in part is a consequence of exposure to circulating testosterone. As each
hemisphere is responsible for the opposite side of the body, it is the left hemisphere
that controls the right hand and is thus dominant in those who are right handed. The
side controlling the most “dextrous” hand is generally known as the dominant hemi-
sphere. This is a rather misleading term as the hemisphere does not dominate the other
hemisphere in any meaningful way, though it is certainly different in terms of some of
its other abilities. Linguistic function provides a particularly dramatic example. In 95
percent of right handers, linguistic function is largely confined to the left hemisphere,
that is, it is only on this side that one finds Wernicke’s and Broca’s areas (Springer and
Deutsch 1997). The situation for left handers is not so clear. Seventy percent show this
230 Chapter Seven

pattern; however, in the remaining 30 percent, language is bilaterally represented. The

lateralization of linguistic functions obviously has major implications for the cortex. If
it is damaged in adulthood, for example, as the result of a stroke (a loss of blood sup-
ply to part of the brain), the ability of the opposite hemisphere to compensate for lost
language function is extremely limited. The one aspect of language that is represented
in the nondominant hemisphere is prosody, that is, the inflection or lilt of speech. This
not only makes speech interesting to listen to, it also imparts meaning. Consider the
short-word sentence “This is your book.” Switching the stress from one word to an-
other produces four different statements, or, by giving the sentence a rising inflection,
four different questions. So prosody provides eight shades of meaning. This may be
linked to the observation that some aspects of music analysis, in nonmusicians at least,
appear to be largely carried out in the auditory association cortex on the nondomi-
nant (usually right) hemisphere. In more general terms it has been suggested that the
dominant hemisphere is involved particularly with serial analytical functions, while
the nondominant hemisphere deals with more holistic, integrative functions, such as
visuospatial operations like face recognition. Indeed, it has further been proposed that
language and visuospatial tasks compete for cortical space (i.e., for computational
resources), as in normal individuals who are blindfolded for a number of days, lin-
guistic function starts to invade the visual areas of the cortex. This underlines the fact
that the cortex is a dynamic and competitive environment. This is seen most starkly
when a region of cortex loses its input or target, for example, when a finger or limb
is amputated. The areas of the primary sensory and motor cortices associated with the
missing region do not cease operation; instead, they are rapidly invaded by links from
adjacent areas. As a result, stimulating a limb stump can produce a phantom sensation
of the missing region because the brain still has it assigned to the amputated part. On
a less draconian scale, it is this competition for resources that allows us to develop
new skills.
In the music world today, both genders are well represented, so in studies of the cen-
tral nervous system of musicians, it is important to consider whether there are signifi-
cant differences between the sexes in cognitive and other abilities or in brain structure.
There is in fact ample evidence for this (Kimura 1999), which supports the general
impression that, as far as motor skills are concerned, men are better at targeting (throw-
ing) and interception (catching). On the other hand, women are in general superior at
fine manipulations of the fingers and the copying of hand postures. In the area of spatial
skills, men are superior at map reading and when finding their way in this fashion tend
to use direction and distance as guiding parameters, while women navigate more with
reference to landmarks. Men are also superior at complex visualization tasks such as
imagining the effects of rotating two- and three-dimensional objects. Women, on the
other hand, are better at remembering the relative positions of objects. When it comes
to mathematics, women tend to be better at calculation, while men are better at apply-
ing reasoning or math to real problems. In the realms of perception, women outperform
men in reading the emotional content of facial expression. They are also more fluent
verbally, though there is no gender difference in vocabulary size.
Studies comparing the size and morphology of particular brain regions in different
groups of people can be difficult to carry out due to the degree of individual variation,
 The Structure and Organization of the Brain 231

which may reflect underlying factors not detected by the experimenters. Nonetheless,
there is some evidence that structural differences also exist between the brains of men
and women (Cahill 2005). Overall, the male brain is slightly heavier when corrected
for body size, but the cerebellum and parts of the corpus callosum are proportionately
larger in females. It has also been suggested that on average the degree of hemispheric
asymmetry is less in right-handed women than it is in right-handed men. There appear
to be gender differences in the size of some of the nuclei within the hypothalamus
and, as this part of the brain is responsible for hormonal control of the menstrual
cycle, clear functional differences also. This means that in searching for changes in
brain organization that may be brought about by musical training, gender is one of the
factors that must be taken into consideration. We will see from the next chapter that
some studies that have pooled data from the two sexes have reached conclusions that
on more careful examination have not proved equally applicable to both.

REFERENCES

Amunts, K., G. Schlaug, A. Schleicher, H. Steinmetz, A. Dabringhaus, P. E. Roland, and K. Zilles.

1996. Asymmetry in the human motor cortex and handedness. Neuroimage 4:216–222.
Cahill, L. 2005. His brain, her brain. Sci Am 292 (5):40–47.
Kell C. A., K. von Kriegstein, A. Rosler, A. Kleinschmidt, and H. Laufs. 2005. The sensory
cortical representation of the human penis: Revisiting somatotopy in the male homunculus.
J Neurosci 25:5984–5987.
Kimura, D. 1999. Sex and cognition. Cambridge, Mass.: MIT Press.
Sacks, O. W. 1995. The case of the colorblind painter. In An anthropologist on Mars: Seven
paradoxical tales. New York: Knopf.
———. 1998. The man who mistook his wife for a hat and other clinical tales. 1st Touchstone
ed. New York: Simon & Schuster.
Penfield, W., and E. Boldrey. 1937. Somatic motor and sensory representation in the cerebral
cortex of man as studied by electrical stimulation. Brain 60:389–443.
Penfield, W., and T. Rasmussen. 1950. The cerebral cortex of man; a clinical study of localiza-
tion of function. New York: Macmillan.
Springer, S. P., and G. Deutsch. 1997. Left brain, right brain: Perspectives from cognitive
neuroscience. 5th ed. New York: Freeman.
Zeki, S. 1992. The visual image in mind and brain. Sci Am 267 (3):68–76.
 Chapter Eight

How the Performance of

Music Affects the Brain

With the knowledge we have gained from the preceding chapter, we can now move on
to consider how the performance of music alters the brain and its responses. We will
deal here mainly with the regions of the brain involved in the physical performance
of music, concentrating particularly on sensory function and on the control and coor-
dination of muscle activity. This will lead us to a consideration of how expert skills
are developed during practice. Finally, we will examine a disturbing and intractable
condition called focal dystonia, which sometimes drastically curtails the professional
life of musicians in midcareer and which appears to be a consequence of maladaptive
changes in brain organization. The response of the brain to musical sound will be dealt
with in the next chapter.

DIFFERENCES IN BRAIN STRUCTURE

BETWEEN MUSICIANS AND NONMUSICIANS

Just as a muscle becomes larger and stronger with use and the bone to which it is
attached grows thicker under continuous loading, the circuitry of the central nervous
system adapts to the demands that are made upon it. The ability of the brain to be
molded by experience is called plasticity and is greatest during childhood, when it
may be capable of compensating almost completely for damage that, if incurred later
in life, would cause catastrophic loss of function. Much of this plasticity we take for
granted. The ability to lay down new memories and acquire new skills throughout life
is a necessity of normal existence. Musicianship is a particularly potent driving force
for plasticity. One would hardly consider starting to train a child under school age in
the skills required for most other occupations; however, it is regarded as perfectly so-
cially acceptable (and even desirable) to do so in an activity that has the higher cultural
overtones possessed by certain forms of music. Quite young children may already be
spending regular periods each day developing the very specialized perceptual and
motor abilities that musical performance demands, and the exceptional standard that
some of these children achieve by their teenage years is to a considerable degree a
testament to the early plasticity of the central nervous system.

232
 How the Performance of Music Affects the Brain 233

The advent of techniques for scanning the bodies of living subjects has opened up
the possibility of analyzing the structure and activity of the brain while it is carrying
out a variety of cognitive and motor tasks. The large and very expensive machines
used for this are increasingly widely available in hospitals and research centers, where
they have rapidly taken on a major role, not only in the diagnosis of disease, but also
in the study of normal brain function. The most widely used scanning technique for
studying the effect of music-related activity on the brain is magnetic resonance imag-
ing (MRI). This can generate images of a series of virtual slices through the brain or
combine them to produce a three-dimensional reconstruction of its entirety. When it is
used to reveal patterns of brain activity, it is known as functional MRI (fMRI). Blood
flow is enhanced in active areas of brain tissue in order to bring in more oxygen and
nutrients to support the higher levels of metabolism they require. With fMRI it is pos-
sible to obtain a blood oxygen level–dependent (BOLD) signal because the scanner
can distinguish between oxygenated and deoxygenated hemoglobin. The informa-
tion this yields allows a detailed analysis of cortical and subcortical activity patterns
underlying musical tasks, as well as making it possible to compare these patterns in
different individuals (e.g., to compare brain activity in musicians and nonmusicians).
One limitation of fMRI is that it cannot track very rapid changes in activity, as it
takes several seconds for the data in a single scan to be collected. To record the fleet-
ing changes typical of normal brain function, it is necessary to measure the electrical
activity of the brain more directly. At one time the principal method of doing this was
electroencephalography, but today this has been largely superseded by magnetoen-
cephalography (MEG). Sensors placed close to the scalp monitor the magnetic fields
that are generated by electrical activity. Computer analysis of the data can be used to
determine the likely location of the source of the signals with a fair degree of accu-
racy. The spatial resolution is greatest for sources near the surface of the brain (i.e., the
cortex) but falls off when the center of activity lies deep below the surface. It is there-
fore less useful for signals arising from the basal ganglia, for example. There is also
the possibility that by using this type of analysis, two separate foci will be interpreted
as a single source lying in an intermediate position. The increased availability of fMRI
and MEG has stimulated a flurry of scientific activity; however, the complexity of the
brain and its circuitry means that interpreting the results is far from straightforward.
It is important to avoid leaping to conclusions from patterns of activity that we do not
yet properly comprehend. In most cases it is not possible to attribute a single function
to a given brain region (even if we were able to define clearly the discrete functional
elements of the brain). It is becoming increasingly clear that the same region may be
involved in the processing of many different types of information and in most cases
it will be some time before we understand what fundamental aspects of information
processing underlie this.
The same caveat applies to the analysis of differences in the structure of the brain
between individuals. The foundation for our understanding of the functional geogra-
phy of the brain was laid in the nineteenth century from observations of the conse-
quences of traumatic head injuries in survivors of wars and industrial accidents, or of
natural accidents such as strokes (a loss of the blood supply to regions of the brain
due to damage to or blockage of blood vessels). The results were collated to produce
234 Chapter Eight

a crude map of cerebral function that still forms the basis of most standard textbook
descriptions today. We are only now beginning to learn that these are often only a
partial reflection of the physiological properties of the brain regions concerned. In
addition, some functions are not strongly localized to a discrete location in the cortex,
but are instead the properties of diffuse networks covering large areas. Many years
were spent in a fruitless search for the location of the engram—the imprint of memory.
This is now regarded as a distributed function, probably residing within a network
in which many memories are superimposed. Each memory may be represented as a
different pattern of activity within the network, in the same way that multiple views
of an image can be stored within a holographic crystal. Nevertheless, there are many
features of brain maps that have stood the test of time, though there remain numerous
pitfalls for the unwary.
Though the “science” of phrenology, which purported to be able to deduce char-
acter from the shape of the head, has long been discredited, there is some evidence
that individual experience can leave a physical imprint on the structure of the brain,
particularly if the shaping influences are present early enough in life. Modifications
in brain circuitry that occur later, while highly significant for brain function, do not in
general result in gross anatomical changes. As a result, when searching for anatomical
consequences of musical training, it is usually necessary to treat individuals whose
training started before the age of about seven as a separate group from those who,
though they may have spent many years or even decades playing intensively, began
their training at a more advanced age. Most of the plasticity observed to date is in the
cortex and its connections. The major anatomical features of the cortex observable to
the naked eye are the ridges (gyri) and grooves (sulci) on its surface, and comparisons
between musicians and nonmusicians based on these features depend on there being
a consistent relationship between topography and cortical function. While this does
seem valid in broad terms, one should nevertheless be cautious, particularly when
only subtle differences are reported, as the difficulties of defining the boundaries of
cortical areas could render them meaningless. Another factor that must be taken into
consideration is that there is considerable variation in brain size between individuals,
and so the volumes or areas to be compared must be scaled according to the overall
dimensions of the brain.
Though most changes in the functional organization of the brain are not reflected
in modifications to its gross anatomy, they can nevertheless be demonstrated with
the imaging methods we have just encountered. Of course, when a music-related task
is being studied, not all (or even most) of the activity in the brain will be associated
with music alone. It is therefore common practice to monitor the signal from a neutral
activity not associated with music and to subtract this from the experimental pattern in
an attempt to isolate the areas that are specifically linked to the task in question. This
is called masking. Another technique is to compare activity patterns when subjects
are asked explicitly or implicitly to analyze different aspects of patterned sound. A
third method is to compare activity between different groups of subjects carrying out
the same task. For example, one can subtract the signals obtained from expert musi-
cians with those from amateurs or nonmusicians to identify the effects of training. The
range of different subtraction methods used, together with the consequences of other
 How the Performance of Music Affects the Brain 235

differences in experimental design, can make comparisons between fMRI studies of

the same phenomena frustratingly difficult; as a result, the published literature is often
rather confusing. If this was not enough, there are additional constraints imposed by
the scanners themselves. The first is that they are noisy machines, which can create
difficulties for investigations of auditory processing. The second is that the space
available for the subject within the machines is very restricted, which means that only
small movements can be made. In addition, it is important that the head remain as sta-
tionary as possible. Finally, because these machines use very strong magnetic fields,
metal objects cannot be placed near them. All of these place practical limitations on
the activities that can be studied.
Musicians are no more a homogeneous group than any other human population, and
the degree of individual variation means that conclusions about how the brain may be
modified by musical experience can only be drawn from comparisons of large popula-
tions of musicians (amateur and professional) and nonmusicians. It is also important to
consider the possibility that the differences observed between the brains of musicians
and nonmusicians may not just be a consequence of musical experience, but may also
reflect inherited characteristics that predispose some individuals to success in music.
For example, one study has suggested that male musicians have lower levels of tes-
tosterone in their bloodstream than male nonmusicians, while female musicians have
higher levels than female nonmusicians (Hassler 2000). At the present time the con-
sequences of this are not clear, but during embryonic development testosterone levels
have an effect on structural and functional asymmetry in the cortex of the forebrain
(i.e., on hemispheric dominance). If these levels are elevated at crucial times or if the
brain has a high sensitivity to testosterone, the development of the left hemisphere
may be held back, while that of the right hemisphere is enhanced. This shifts the bal-
ance between the ability to process verbal and nonverbal information and may favor
the emergence of musical and other abilities. Those carrying out morphological brain
studies of musicians must therefore try to distinguish features that might have arisen
in this way from those that reflect the result of musical experience. This may not be an
easy call, and often relies on reasoning based on rather circumstantial evidence.
Musical perception involves the analysis of many different parameters such as
melody, rhythm, harmony, and timbre, and how our brains have been shaped to
respond to these depends to some extent on expectations based on our previous ex-
perience. This causes us to formulate certain rules of syntax and to react particularly
strongly to events that disobey these rules, for example discords or unexpected chord
progressions. The rules, and hence the expectations, are culture dependent, and even
within the Western musical tradition, what is harmonically acceptable has changed
throughout history. Consequently, some of the details of the observations made on
the functional organization of the brains of musicians raised in the Western classical
tradition (who are the main subjects of published studies to date) may not be entirely
applicable to all musical cultures.
We will now turn our attention to the areas of the brain that are involved in the
acquisition of the skills necessary for musical performance. In each case, we will first
consider the general organization of the system, and then how this is modified by musi-
cal training. The music-related changes discussed are summarized in table 8-1. We will
Table 8-1. Changes in Brain Structure Related to Musical Experience
Brain Region Change
Heschl’s gyrus (primary Increased gray matter
auditory cortex) volume up to 130%
Left PT Greater leftward
asymmetry (smaller
right PT)
Primary sensory cortex Greater left-hand map
area (right cortex)
Primary motor cortex Intrasulcal length
greater in right-hand
region
Corpus callosum Enlargement especially
in anterior part
Corpus callosum Increased organization/
myelination?
Broca’s area Increased gray matter
Cerebellum Increased gray matter
volume
PT = planum temporale
? = some contradictory evidence
* = few of the musicians in this study started playing at an early age, so correlations unlikely
Instrument Correlation
Not specified Musical aptitude/experience
(professional > amateur >
nonmusician)
Mainly string and Absolute pitch, early
keyboard players musical training
String players Early musical training?
Keyboard players Early musical training
Keyboard and string Musical experience
players
Keyboard players Musical training tested
Various orchestral Total years of playing, not
players early musical experience*
Keyboard players Lifelong practice intensity,
not early musical
experience or total
playing years
Gender Factors Source
Not known Schneider et al. 2002
Males only Schlaug et al. 1995
Zatorre 1998
Luders et al. 2004
Not known Elbert et al. 1995
Hashimoto et al. 2004
Only males tested Amunts et al. 1997
Males only? Schlaug et al. 1995
Ozturk et al. 2002
Lee et al. 2003
Only males Bengtsson et al. 2005
Only males tested Sluming et al. 2002
Males only Schlaug 2001
Hutchinson et al. 2003
Figure 8-1. Cortical and subcortical motor areas of the brain. A, B. The primary motor area of
the cortex is driven by input from the supplementary and premotor areas. These make important
contributions to the planning and initiation of movements and the establishment and storage of
learned motor patterns. Sensory feedback from primary somatosensory and sensory association
areas is necessary for these operations. Broca’s area is a speech motor area (left hemisphere only)
that not only contributes to the control of the muscles used in speech, but is also involved in the
analysis of grammar in language. C. The accurate execution of movement is also dependent on
processes that take place in the spinal cord and in regions of the brain outside the cortex. These
include the thalamus and elements of the basal ganglia that lie in the forebrain and midbrain. The
cerebellum is involved in the maintenance of equilibrium and plays a central role in motor learn-
ing. It receives sensory information from the muscles and the balance organs of the ears as well
as input from the cortex. In order to make it easier to come to terms with the diverse functions of
these different areas, they have been given metaphorical subtitles here and in figure 8-2.
238 Chapter Eight

start with a consideration of the regions of the brain that receive sensory information
from the body, in particular from the hands, and the motor areas that control and co-
ordinate the actions of the muscles used in performance. For reasons that will become
apparent, these are among the most extensively studied systems from the viewpoint of
music-related plasticity.

SENSORY AREAS OF THE BRAIN

In the first chapter we discussed the different types of sensation that sensory nerve
cells carry. From this it will be remembered that information coming from the skin
is said to be exteroceptive because it monitors events that originate in the external
world, such as temperature, touch, and vibration. We also receive information from
the muscles, tendons, and joints that tells us about the position of limbs and the load-
ing of muscles. This is called proprioceptive information and is very important for the
control of movement. Both of these types of sensation are central to performance.
The axons of exteroceptive sensory nerve cells enter the spinal cord and contact
interneurons that ascend to the brain stem before terminating in the thalamus on the
opposite side of the brain. Here they contact other nerve cells that send their axons
to the primary sensory cortex. The left sensory cortex therefore receives information
from the right side of the body and vice versa. The primary sensory cortex is a narrow
strip of gray matter that runs parallel to the central sulcus (fig. 8-1; see also previous
chapter). Projected onto this strip is a map of the surface of the body called the sensory
homunculus (literally, the “little sensory man”). In figure 7-6C, he is shown in three
dimensions, from which it can be seen that though the map is complete, it is distorted
so that regions such as the hands, lips, and tongue are disproportionately large. The
areas that are most strongly represented in the map are those with the highest concen-
tration of sensory nerve endings. The sensory homunculus does not therefore repre-
sent how we see ourselves, but what it feels like to be us. We are mostly unaware of
this distortion, but if you have ever had a filling fall out of a tooth you will know that
when you explore the hole with your tongue, the hole seems much larger than it looks
in the mirror. This is because the tongue is overrepresented on the sensory map due
to the high concentration of sensory endings on its surface. The densely distributed
sensory endings in the skin on different parts of the body can be compared by using a
pair of dividers to measure how far apart two points of contact must be to be perceived
subjectively as separate. On the fingertips, the lips, or the tongue, this is on the order
of a couple of millimeters, but on the palm it increases to a centimeter and on the arm
to four centimeters, while on the back it can be as much as 40 centimeters (try it!).
The proprioceptive sensory information from the muscles and tendons also con-
tributes to the map in the sensory cortex and is represented in a strip of cortex that
runs parallel to the band receiving information from the exteroceptors. Proprioceptive
information goes not only to the cortex, but also to the cerebellum where it plays a
role in the feedback control of movement. However, instead of being passed to the
half of the cerebellum on the opposite side of the brain as happens with the cortex, it
runs to the half that lies on the same side of the body as the muscle, tendon, or joint
 How the Performance of Music Affects the Brain 239

from which the information originates. Sensory information from the left side of the
body therefore generates activity in the left cerebellum but the right sensory cortex.
This has to be borne in mind when interpreting the results of research into patterns of
brain activity produced by musical activities.
Sensory feedback is clearly important for monitoring the physical processes in-
volved in the playing of a musical instrument. Sensory signals from the skin on fingers
provide information about the position of the pads on the keys, the vibration of strings,
and the effects of movements that may be used to generate vibrato. Signals from the
muscles, tendons, and joints provide information about the position assumed by the
hands and arms and the forces being generated by movement. For many instruments,
one of the key early skills that must be developed is to be able to place the fingers
in the correct position without having to look at them. For wind players, tactile and
vibratory sensations from the lips and sensory input from the muscles of the embou-
chure are important for regulating the pitch and the quality of the sound.

Music-Induced Plasticity
The sensory maps in the brain are not immutable but can be remolded in the light of
sensory experience. Though the changes in the map may be radical, this does not usu-
ally produce any change in the dimensions of the sulci or gyri of the primary sensory
cortex, and so expansion of the representation of one region of the body must be at the
expense of that of another. We have seen that if a limb or finger is lost, the region of
the cortex that received information from it is quickly taken over by adjacent regions
of the body. The map can also be modified in the opposite direction. Some people
are born with their fingers fused together, and this is reflected in the sensory cortex
where these fingers are represented as a single structure. If the fingers are surgically
separated, the map adapts to this by dividing the representation into two distinct areas.
These effects are rapid, taking place over a period of weeks and demonstrating that the
map is highly dynamic, being maintained by constant competition between sensory
signals from adjacent regions of the body surface. Changes in the cortical map can
take place much faster than this, however, even over a period of only a few minutes
(Noppeney et al. 1999). This implies that several mechanisms for plasticity exist that
operate over different time scales. The short-term changes are probably the effect of
alterations in activity within circuits, which act to block some connections and en-
hance others. The long-term changes, on the other hand, are likely to be the result of
alterations to the structure of the circuit involving growth of nerve cell branches and
the forming of new connections.
Two studies have been carried out on string players to see if it is possible to dem-
onstrate changes to the representation of the hand that are consistent with theoretical
expectations. Though the results differed somewhat in detail, both reported an en-
larged cortical representation of the left hand compared to that of the right (Elbert et
al. 1995; Hashimoto et al. 2004). The cortical representations of the left hand of the
musicians were also proportionately larger than those in a group of nonmusicians.
Both studies also revealed a shift in the position of the map within the cortex, though
not in the same direction. This indicated that it was encroaching into areas that would
240 Chapter Eight

normally map other regions of the body surface. However, while one of the studies
(Elbert et al. 1995) found that the magnitude of the changes appeared to be correlated
with the age at which musical training began, the other (Hashimoto et al. 2004) did
not. Though currently unresolved, this discrepancy probably reflects the small number
of musicians studied in each group and methodological differences in the experiments.
In general, the degree of modification of the sensory maps does not appear to be solely
a function of the pattern of sensory stimulation. It is not the repetition of the activity
per se that leads to enhancement of the associated area of the map. The effect depends
significantly on whether the attention of the subject is directed to the finger or fingers
being stimulated (Noppeney et al. 1999). This, of course, is likely to be the case in
musicians perfecting their playing technique and is a feature of what we will encoun-
ter as “deliberate” practice later in this chapter.

MOTOR AREAS OF THE BRAIN

The roles and interrelationships of the many areas within the brain that control the
muscles are complex; to make it easier to understand what I will do is use a number
of musical analogies (figs. 8-1, 8-2). While these can never be exact or truly reflect
the full functionality of this system (to which a considerable proportion of the brain’s

Figure 8-2. This diagram summarizes some of the relationships between the main regions of the
cortex involved in motor activity shown in figure 8-1.
 How the Performance of Music Affects the Brain 241

resources are devoted), I hope that by this means you will be better able to grasp the
significance of the principal components of this intricate system.

Motor Areas Inside the Cortex

Running parallel with the primary sensory cortex and lying immediately in front of it
on the other side of the central sulcus is the primary motor cortex (Kandel, Schwartz,
and Jessell 2000). Like the primary sensory cortex, it also contains a map of the
body—this time of the muscles on the opposite side. The motor map lies roughly in
register with the sensory map, with the foot and leg lying close to the upper surface
of the cortex, the trunk on the upper surface, the arm and hand on the lateral surface,
and below that the face (fig. 7-6). As with the sensory representation, the motor map is
distorted because small muscles that perform intricate or finely controlled movements
(such as those moving the fingers or the muscles of facial expression that contribute
to the embouchure) have a larger area of the cortex devoted to their control than pos-
tural muscles such as those of the thigh, which, though much larger, do not require
such precise control. This is related to differences in the number of motor units within
the muscles. A motor unit is the group of muscle fibers controlled by a single motor
neuron (see chapter 1). If the force generated by a muscle is to be controlled very
precisely, it must contain a large number of small motor units and will need more
cortical neurons to control it than a muscle that has a smaller number of motor units.
Nerve cells in the primary motor cortex run down through the brain and into the spinal
cord and in many cases make direct contact with pools of motor neurons that supply
the muscles. These motor neuron pools may lie either in the brain stem (if the motor
neurons run in cranial nerves supplying muscles in the head and neck) or in the spinal
cord, so the axons of the neurons in the motor cortex are often very long. The major
bundle of axons that runs down from the motor cortex to the spinal cord is called the
pyramidal tract and is particularly important for the control of fine hand movements. If
it is damaged, motor control of much of the body may recover at least to some extent,
but individual control of the fingers is lost.
The structure of the motor map was originally thought to be quite similar to the
sensory map. If a brief electrical stimulation was applied to a single point on the pri-
mary motor cortex, a well-defined discrete movement was produced that appeared to
be the result of activity in a single muscle. This was first observed in humans whose
brains were exposed during surgery, but similar results can now be obtained in nor-
mal awake individuals by using magnetic field stimulation, which does not involve
broaching the skull. This led to the notion that the map was a simple representation of
individual muscles. To use a musical analogy, it was seen as representing a keyboard
on which the rest of the brain could play. Pressing down a key activates a mechanism
that leads to the sounding of a single note or, in this case, a single muscle. In this anal-
ogy, the motor neurons in the spinal cord act like the hammers of the piano and are
responsible for sounding the strings (the muscles). However, this notion has proved
to be a considerable oversimplification. Though the motor map initially appeared to
run continuously from the toes to the fingers, when examined closely it was found to
be fragmented. The nerve cells controlling a particular muscle do not all lie precisely
242 Chapter Eight

in the same place but are scattered across a small area of the motor cortex. As a re-
sult, while the area of motor cortex controlling the hand is relatively easy to define,
the representations of the individual muscles moving the fingers are to some extent
intermingled. Furthermore, it now appears that single nerve cells in the primary motor
cortex may supply several muscles that may even control more than one joint. Using
pulses of electrical stimulation that are longer or more intense than those needed to
produce twitches in single muscles evokes complex but well coordinated movements.
It transpires that a given cortical neuron may drive a muscle or set of muscles only
during one particular movement and remain silent when the same muscles are used in
a different context. This has led to the idea that the cortical map may in fact represent
movement trajectories (Graziano et al. 2002). To return to our musical analogy, the
pressing of a key results not in the striking of a single note, but in the sounding of a
chord. To produce a different chord, another key must be pressed, even though several
of the same notes are present in both.
Activity in the primary motor cortex is driven or influenced by connections from a
number of other cortical regions (figs. 8-1, 8-2). Not surprisingly, one of these is the
primary sensory cortex. Additional connections arise from sensory association areas
such as the superior parietal region, which sits behind the primary sensory cortex and
plays a role in integrating the sensory information that is used in the planning of motor
activity. In addition to these sensory streams, the primary motor cortex also receives
inputs from several other motor areas that lie just in front of it (figs. 8-1, 8-2). These are
principally the supplementary motor cortex (which lies on the top or superior surface of
the hemisphere) and the premotor areas (which sit in front of the primary motor cortex).
Both also have connections to the motor pools in the brain stem and spinal cord, so they
can act on these directly as well as through the motor cortex. Like the motor cortex,
they receive information from sensory association areas. Information from the primary
visual cortex reaches the premotor cortex via the visual association areas along two
processing paths. The more ventral stream carries information on the shape and position
of objects and is used to direct reaching and grasping behavior. The more dorsal stream
is active when visual and other sensory signals trigger a movement but do not guide it.
The supplementary area, by contrast, is concerned with movement that is self-generated
rather than triggered by external cues. Into this category fall the movements required
for playing an instrument and the finger-tapping tasks discussed below in the context
of motor learning. It also controls sequences of movements replayed from memory. We
can therefore consider it as a “score library” for movement, though perhaps more of
short “phrases” than of complete “pieces.” When a new set of movements is first being
learned, an area of the cortex that lies just in front of the supplementary motor cortex
(imaginatively named the presupplementary motor area) is active. Though connected to
it, the supplementary motor cortex is silent during this initial period, but once learning
is complete it becomes active when the sequences are reenacted. With greater practice,
the replaying of the movement sequences becomes fully automatic. The supplementary
motor cortex may then fall relatively quiet again, and the activity becomes largely con-
fined to the primary motor cortex. This drift in cortical activation during learning will
be encountered again when we discuss the contribution of the different motor areas of
the brain in the context of musical experience (fig. 8-3).
 How the Performance of Music Affects the Brain 243

Motor Areas Outside the Cortex

In addition to the spinal cord, motor control involves many areas of the brain that
lie outside the cortex (figs. 8-1C, 8-2). Substantial contributions are made by large
masses of gray matter deep within the forebrain and midbrain that are collectively
known as the basal ganglia, and by the cerebellum. The separate elements of the basal
ganglia are linked to each other and to the various motor areas of the cortex through
a series of loops that involve part of the thalamus. These pathways modulate the
motor activity of the cortex and help to ensure that individual movements generated
run smoothly into one another. Their corporate action can be regarded as being like
that of an orchestral conductor. Not surprisingly, therefore, they are important in the
learning of new motor tasks. Their role is perhaps best appreciated by observing what
goes wrong when they cease to function effectively. Basal ganglion problems fall into
two broad categories, those in which there is too little movement and those in which
there is too much. A familiar example of the former is Parkinson’s disease. Patients
with this condition may find it difficult to initiate a movement and once started, a
sequence of movements may cease partway through. The movements themselves may
be abnormal; for example, when walking, the steps taken may be short and shuffling.
At the other extreme are conditions where movements are exaggerated or appear
spontaneously against the wishes of the patient. Movements in this category can be de-
scribed as choreas (dancelike jerks—the word has the same origin as choreography).
Huntingdon’s chorea is a genetically inherited condition affecting the basal ganglia.
Other abnormal movements are writhings (athetosis) and tics (twitching of single
muscles or small groups of muscles).
The cerebellum is another major player in the control of movement. It is a complex
structure that contains more than a quarter of all the nerve cells in the entire central
nervous system. It receives information from the organs of balance within the inner
ear (the vestibular system) and so one of its main functions is to maintain our equilib-
rium through the action of the muscles of the trunk and the big postural muscles at the
base of the limbs, particularly the legs. As we stand upright on two rather small feet,
maintaining our balance requires the constant intervention of the cerebellum, though
we are almost entirely unaware of this. When a movement is initiated by the motor
cortex, it sends information about what is intended to the cerebellum, which then com-
pares this with what is actually taking place. We can consider this role to resemble that
of the répétiteur who ensures that the words that are spoken or sung by the performers
on stage match those in the script. The telemetry that allows it to do this is provided
by the proprioceptors in the muscles and tendons, some of the sensory neurons sup-
plying the skin (particularly the soles of the feet), the eyes, and the vestibular system.
It then generates a signal that directs the motor cortex to make corrections to the
movement even while it is still taking place. This ensures that the outcome matches
the expectation, which is particularly important during the acquisition of motor skills.
For example, when we learn to ride a bike, the challenge of maintaining balance at
first seems insuperable, but rapidly becomes automatic. The importance of the cer-
ebellum in everyday life is easily seen when its normal functions are compromised by
overindulgence in alcohol!
Figure 8-3
 How the Performance of Music Affects the Brain 245

Figure 8-3. This figure highlights the sensory and motor areas of the cortex that are involved
in hand control, for example, in keyboard or string playing. It also shows their changing pattern
of activation during skill learning. A. The regions of the cortex most involved in the control and
coordination of muscle activity are shaded. Their major contributions and connections are sum-
marized in figure 8-2. B. The regions particularly involved during keyboard or string playing are
highlighted. Activity is focused in regions that control the hand. C. During mental practice, the
pattern of activity is similar to that for playing except that the somatosensory cortex is not receiv-
ing information from the fingers, and the region of the motor cortex controlling the hand is silent.
Panels D–F schematically illustrate the changes in cortical activation as performance skills are
first learned then perfected. While new motor patterns are being laid down or when a piece is
played for the first time, there is strong activity in the presupplementary cortex. As the movement
sequences are repeated during learning, activity soon shifts to the premotor cortex (E). When
playing becomes fully automatic and a high level of competence is achieved, premotor activity is
greatly reduced, while activity in the primary motor and sensory cortex becomes intensified and
more tightly focused within the hand area.

Within the spinal cord, the pools of motor neurons that supply individual muscles
lie in the ventral horn of the gray matter (see chapter 1). Sensory information from
the muscles, tendons, and skin flows into the cord and though, as we have seen, it is
sent up to the brain, branches from the sensory neurons also end in the cord itself. As
a result, all of the circuitry needed to control some simple forms of motor behavior is
found here. Limb reflexes, for example, are mainly spinal. Tapping the large tendon
below the kneecap causes a knee-jerk reaction—a spinal reflex over which we have
no voluntary control. When we walk, as we swing one arm or leg forward, the other
moves back. The nerve circuitry responsible is again confined in the spinal cord.
Spinal circuitry also reduces interference between antagonistic muscles. If we use the
triceps muscle to straighten the arm, the muscles that flex it relax. The presence of
spinal reflexes simplifies the task required by the motor cortex in controlling many
types of movement, though it also retains the option to override the spinal circuitry if
this is appropriate.

PERFORMANCE-RELATED BRAIN ACTIVITY IN

AMATEUR AND PROFESSIONAL MUSICIANS

Patterns of brain activity differ considerably between professional and amateur musi-
cians, even during the playing of quite simple pieces of music. In one study, profes-
sional violinists playing around thirty hours a week and with an average of thirty
years’ experience were compared with amateurs who played only one hour a week and
had about ten years of training (Lotze et al. 2003). Their brain activity was recorded
using fMRI while they performed the left-hand finger movements required to play a
short extract of a Mozart concerto. Because the experiment was carried out within the
limited confines of a brain scanner, the piece was memorized and the finger move-
ments of the left hand made without the instrument. As would be expected, there were
many similarities between the brain responses of the professionals and the amateurs;
both showed activity in primary sensory and primary motor areas that represented the
hand. However, the activity in the professional group was much more tightly focused
246 Chapter Eight

and in the primary motor cortex was more intense and confined to the right side of
the brain (the side that controls the left hand), whereas in the amateurs it was present
on both sides. The stronger signal from the hand area of the cortex in the professional
group may be a reflection of an increase in its cortical representation (Schlaug 2001;
Amunts et al. 1997). The supplementary and premotor regions of the cortex were also
active in both groups, though more so in the amateurs. These, together with some
other frontal areas, and the left side of the cerebellum (as expected on the same side
as the active hand), which were also active in amateurs, are involved in the acquisition
of complex motor skills before they become fully automatic. In professional players,
therefore, it appears that many of the complex motor programs required for play-
ing have become fully integrated and refined so that they arise fully formed directly
from the primary motor cortex. Activity in the basal ganglia, which is often seen at
an early stage in the formation of motor programs, was found only in the amateurs,
again reflecting their lower level of proficiency. This may also underlie a greater
level of activity in the right side of the cerebellum in the amateurs. By contrast, in the
professionals, particularly those who started their training in early life, there was an
increased level of activity in a small region of the cerebellum on the same side as the
active left hand. This may be correlated with the observed structural changes seen in
the cerebellum of musicians (Schlaug 2001; Gaser and Schlaug 2003).

MUSIC-INDUCED STRUCTURAL CHANGES

IN THE MOTOR AREAS OF THE BRAIN

Because of the complexity of the demands made on their manual dexterity, keyboard
players are among the most often studied in the search for changes in brain structure
and function related to musical performance. The primary motor and sensory areas
of the cortex are obvious places to search for such modifications because they con-
tain topographical body maps in which distortions, should they exist, will be readily
detectable. Most people show a greater dexterity with one hand or the other; that is,
they are either right or left handed, and brain imaging studies have revealed that this
is reflected in the depth of the central sulcus that lies along the edge of the primary
motor cortex on the opposite side of the brain (Amunts et al. 1996). Because there
are more cells in this sheet of gray matter on the side controlling the dominant hand,
it bulges out more, making the sulcus deeper. As the strip of gray matter that makes
up the primary motor cortex contains a map of the body in terms of muscles or
movement, only part of it is concerned with control of the arm and hand. Left/right
asymmetry is seen in this region only, and not in the region lying immediately below,
which controls the muscles of the face. The playing of keyboard instruments requires
an almost equal dexterity in both hands, so it can be hypothesized that the difference
between the depth of the central sulcus on the left and right side should be less in
keyboard players than in the general population. This indeed appears to be the case, at
least for male players (who are the only group studied so far), and can be attributed to
the sulcus being deeper in the pianists on the side controlling the nondominant hand
(Amunts et al. 1997; Schlaug 2001). This anatomical observation correlates with a
 How the Performance of Music Affects the Brain 247

greater symmetry in finger dexterity in the two hands in pianists (Jancke, Schlaug,
and Steinmetz 1997). The depth of the sulcus on both sides of the brain of keyboard
players shows some correlation with the age at which the musicians started to learn to
play. There is no correlation with the total number of years of playing at the time the
study was carried out, indicating that the effect reflects a plasticity in brain structure
present only early in life.
There is evidence that the extent of the gyrus may also be increased in the hand area
of musicians (fig. 8-4). The region of the primary motor cortex controlling the hand
is quite commonly folded into an omega (Ω) shape. Comparisons of the brains of six-
teen violinists, sixteen pianists, and thirty-two nonplayers (all right handed) revealed
that in the violinists the omega shape tended to be considerably more pronounced in
the right primary motor cortex (which controls the left hand) than on the other side,
which of course controls the bowing hand. In pianists, however, the opposite situation
prevailed, and it was on the left side that the omega shape was more strongly marked
(Bangert and Schlaug 2006). Though the most obvious conclusion is that this is a
result of hand use, it is always possible that it is genetic and that musicians self-select
an appropriate instrument, so long-term studies of brain development in children are
currently in progress to clarify the issue (Schlaug 2006).
The two cerebral hemispheres of the brain are linked by an extensive sheet of nerve
fibers called the corpus callosum, and it might be envisaged that the requirement to
coordinate the two hands in instrumentalists such as pianists could lead to changes in
its structure. The corpus callosum matures slowly. Though it grows most rapidly in the
first decade of life, it continues to increase in size into the midtwenties. The period of

Figure 8-4. Morphological features of the primary motor cortex associated with the complex
hand movements required by musicians. These diagrams, based on observations by Bangert and
Schlaug (2006), illustrate that the hand region of the right (R) primary motor cortex of violin
players is more likely to exhibit a well-defined omega-shaped fold than the left. In pianists the
opposite is true. The shaded area behind the central sulcus (dashed line) represents the hand area
of the primary sensory cortex.
248 Chapter Eight

most rapid growth coincides with the time in early childhood when motor coordination
is developing. Sensory deprivation during this period can result in a reduction in the size
of parts of the corpus callosum. A comparison has been made of the size of the callosum
between a population of musicians and one of nonmusicians in the eighteen-to-thirty-
five age range. The musicians were classically trained keyboard and/or string players
who were either professionals or music students, while the members of the control group
were matched for age and educational status but did not play an instrument and had
no musical training. In two separate studies of different groups of male musicians, the
anterior part of the corpus callosum was found to have a greater area than in the control
groups (Lee, Chen, and Schlaug 2003; Schlaug et al. 1995). This region of the corpus
callosum carries connections between the primary sensory and primary motor areas of
the two hemispheres, as well as between the premotor and supplementary motor regions.
Its larger size in musicians might therefore be associated with the observed changes in
the primary motor cortex already discussed. This is likely to be related to the importance
of bimanual coordination in the playing of many instruments. Another study showed a
higher degree of organization within the corpus callosum of male pianists than in age-
matched controls (Bengtsson et al. 2005). This used an MRI technique called diffusion
tensor imaging. The level of organization and the degree of myelination (as indicated
by a parameter called functional anisotropy) was correlated with the amount of practice
carried out during childhood and adolescence. One physiological manifestation of the
enlarged callosum may be a reduction in the inhibition normally seen between the two
sides of the brain that has been reported in pianists and guitarists (Nordstrom and Butler
2002; Ridding, Brouwer, and Nordstrom 2000). This is an inhibition that occurs in the
primary motor cortex on one side of the brain during activity in the equivalent region
on the opposite side, and it is mediated by nerve fibers in the corpus callosum. It has
been hypothesized that its reduction may improve the ability to coordinate the precise
timing of the movements of the two hands. Another investigation of the corpus callosum
involving only string players confirmed the results obtained by Schlaug and colleagues
and also found evidence of enlargement of the posterior part of the corpus callosum
(Ozturk et al. 2002), which is known to carry fibers that link the auditory centers on
each side of the brain.
In contrast to the situation seen in male musicians, one study has suggested that
female musicians show no increase in the dimensions of the corpus callosum (Lee,
Chen, and Schlaug 2003). Several possible explanations have been put forward to ex-
plain this. One is based on the suggestion that females show less asymmetry between
the hemispheres than males. If the changes in the corpus callosum in males were
related to the reduction in asymmetry of the motor cortex already described, then any
change in females due to bimanual training might be much smaller. Another possible
explanation relates to the higher incidence of absolute pitch seen in the female musi-
cian group than in the male group. Absolute pitch is associated with a greater degree
of lateralization (a greater difference in structure between the two hemispheres) in
the auditory regions, which might be reflected in fewer connections crossing between
them. When averaged across a number of female subjects, any increase in size in the
corpus callosum due to hand training may might be masked by its reduction in those
with absolute pitch. On the other hand, though the study by Ozturk (2002) did not
 How the Performance of Music Affects the Brain 249

distinguish between male and female brains, the majority of his subjects were female
and yet he still found an enlargement in the musician group. As there is considerable
variability in both sexes in the size of the corpus callosum, the question of whether
only male musicians show changes in its architecture deserves further investigation.
Since the middle of the nineteenth century, the cerebellum has been associated with
the control of motor activity and motor learning, a relationship that was reinforced by
studies of brain damage in soldiers during the First World War (Parsons 2003). This is
clearly important in learning to play an instrument and in developing expert levels of
performance. However, our ideas of cerebellar function are changing, and evidence is
accumulating that it also makes a significant contribution to sensory discrimination. This
includes a role in the comprehension and analysis of harmony, melody, and rhythm,
where its activity appears to support the high-level processing of auditory information
that takes place in various regions of the cerebral hemispheres (Parsons 2001). Imaging
studies of the cerebellum in living subjects have revealed that its volume (as a percent-
age of total brain volume) appears greater in male musicians than in male nonmusicians
(Schlaug 2001). Interestingly, unlike some other morphological changes, this is said to
be related not to the age at which music training began but to the intensity of current
long-term practice (Hutchinson et al. 2003). The time scale over which this volume
change might take place is unknown. There is no comparable difference in cerebellar
size between female musicians and nonmusicians; however, it appears that the female
nonmusicians already have a relative cerebellar volume comparable to that of the male
musician group. The computational part of the cerebellum is its thin but highly folded
cortex, but most of its volume is made of white matter, which represents the cabling
that links the cortex to other regions of the brain. However, it appears that the increased
volume of the cerebellum in male musicians is not due to an increase in the white matter
alone. The relative size of the region of the cerebellar cortex that is involved in control
of finger movement in the left hand is also positively correlated with the degree of mu-
sical training (Gaser and Schlaug 2003). As all of the musicians in the study were right
handed, this result, like the observation of the reduced asymmetry between the left and
right motor cortex that we have already discussed, represents an increase in the develop-
ment of brain regions controlling the nondominant hand.

MUSIC-INDUCED CHANGES IN
ASSOCIATION AREAS OF THE BRAIN

Comparisons between musicians and nonmusicians have also been made in cortical
gray matter areas. One region where gray matter volume is enhanced in musicians is
the superior parietal cortex, which lies just behind the primary sensory cortex on the
upper surface of the brain (Gaser and Schlaug 2003). This forms part of the sensory
association area shown in figure 8-1 and has been implicated as one component in a
network that is active in sight-reading (Sergent et al. 1992). As we have seen, this area
is involved in the spatial analysis of visual information, an ability required not only
for guiding hand movements, but also for the interpretation of the position of notes
on the stave of written music. Being connected to the primary sensory cortex and
250 Chapter Eight

the primary, supplementary, and premotor areas, it is strategically placed to control

visually guided precision motor tasks such as the placement of a finger on a string or
key. Another region of the cortex that appears to be enhanced in musicians lies within
Broca’s area (Sluming et al. 2002), which is found in the left frontal cortex just above
the tip of the temporal lobe. This area is traditionally regarded as having a role in the
control of speech motor activity, but recent functional studies suggest that this general
area may also be involved in a number of music-related functions (see chapter 9). It
is, for example, implicated in the analysis of rhythm, but it may also be active during
tasks such as sight-reading, and in score reading in the absence of playing (Parsons
2001), as well as in the recognition of melody and the processing of musical syntax
(e.g., detecting inappropriate chord progressions [Maess et al. 2001]).
The subjects of the study of Broca’s area carried out by Sluming were male musi-
cians in a major British orchestra with an age range of twenty-six to sixty-six years
(Sluming et al. 2002) and were compared with an age-matched control group of non-
musicians. Most other investigations of brain morphology in musicians have focused
on students or young professionals, generally less than thirty years of age, so the
Sluming study provided an interesting opportunity to examine the effect of musical
performance on aspects of brain morphology and function throughout the entire period
of a typical career. For musicians under the age of fifty, the degree of enhancement
of Broca’s area appeared to increase with the number of years of playing. Overall, the
control group showed an age-related decline in total brain volume as well as in the
volume of gray matter within Broca’s area; however, neither of these changes were
seen in the musicians. The subjects of the study were given some cognitive tests: one
of basic reasoning ability, which provided a general assessment of cognitive function,
and one of spatial ability, which therefore tested a cognitive function thought to be
related to sight or score-reading skill. On the general reasoning test, the musicians
and the control group performed similarly, both showing an age-related decline in
performance. In the test of spatial ability, though the performance of the controls
declined here also, the performance of the musicians actually improved with age. It
is well established that constant practice of a challenging intellectual task can help to
maintain particular mental abilities and, over a period of time, stimulate cortical vol-
ume increases in adults. One notable example of this is seen in London taxi drivers,
who must memorize a vast amount of geographical information as a prerequisite for
obtaining their cab license. This is known as “The Knowledge” and it takes several
years to acquire. Most drivers do not start on this momentous task until their late teens,
and many are considerably older. It has been found that gaining The Knowledge is
associated with an increase in the volume of an area of the cortex known as the hip-
pocampus (Maguire et al. 2000), which is responsible for laying down associative
memory, including the memory of spatial locations.

LEARNING TO PLAY MUSIC

Developing the skills to play a musical instrument involves a variety of learning tasks.
First it is necessary to learn how to produce the sound and finger the notes. For a first
 How the Performance of Music Affects the Brain 251

instrument, this generally runs in parallel with learning to read musical notation, and
the two tasks must be brought seamlessly together so that seeing a note or group of
notes on the stave automatically leads the fingers to make the correct movement. The
motor behavior is gradually refined, typically over a period of many years, to allow
an increase in the speed of execution and the level of control so that the sound gener-
ated meets the expectations of the player and audience. Very high levels of control are
required not only for accurate rendition of the music, but also to enable it to be played
with expression—a rather intangible but generally instantly recognizable element of
performance that is realized by subtle manipulation of timing and dynamics. Today,
near perfection in the accuracy of execution is expected by audiences spoiled by ex-
posure to recorded music, though most probably never realize the extent of physical
practice that is required not only to achieve the required standard, but also to maintain
it. In addition, as it is also expected that, for soloists at least, even long pieces will be
performed without the score, it must be reliably committed to memory. Many aspects
of the way in which these faculties are developed are still beyond the scope of current
research, but we will consider some of the more mechanistic aspects of skill acquisi-
tion in order to see whether our present perceptions of the underlying biology can
provide any insight into how this is achieved.

The Mechanism of Motor Learning

The studies that are of most relevance to understanding how the technical ability to
play an instrument or a new piece of music is acquired are based on learning simple
patterns of finger movement. Though these resemble patterns used in the playing of
keyboard and other instruments, they do not match them exactly. In these studies, the
motor skills of musicians and nonmusicians are compared. As the exercises have a
degree of novelty for both groups, what is being examined are not changes associated
with the specific skills demanded by a particular instrument, but the general mecha-
nisms that underlie the greater levels of motor performance that musicians develop.
Several investigations use a task that requires that the pads of different fingers be
brought into contact with the pad of the thumb in a particular sequence. The subjects
are asked to do this as rapidly and as accurately as possible and without looking at the
fingers. The results suggest that we learn to carry out such patterns of movement in
several stages (Karni et al. 1995, 1998). An initial period of fast learning takes place
over a period of minutes, during which there is a significant improvement in perfor-
mance; however, this is followed by a subsequent period of consolidation lasting six to
eight hours that occurs in the absence of the activity. Thus, when the task is repeated
(e.g., the next day) there is a further improvement in performance beyond the final
level achieved at the end of the first practice session. Though a gradual and continuous
improvement in performance is seen in the period following training that occurs dur-
ing waking hours, consolidation of motor learning appears to be dependent on sleep
(Walker et al. 2003). This results in an improvement in both the speed and accuracy
of the learned task (e.g., a finger-tapping sequence), which in the hours following
sleep remains at a plateau in the absence of further training on the following day. If
additional days and nights are allowed to elapse without further training, performance
252 Chapter Eight

continues to improve gradually. Though sleep has been implicated in the consolida-
tion of several forms of memory, the crucial phase of sleep involved appears to vary
depending on the learning task. In the case of motor learning this is what is called
phase 2 sleep, while for visual discriminative tasks it is the slow wave and rapid eye
movement periods that are critical (Stickgold and Walker 2005).
During the next few weeks, daily practice sessions produce additional improve-
ment, but the increments in improvement become progressively smaller until an upper
level of proficiency is reached. To break through this ceiling requires a considerable
increase in effort. Some notions of the mechanisms underlying these processes have
been gleaned from imaging brain activity during this type of learning. When the task
is carried out for the first time, activity is seen in the primary motor cortex of the
hemisphere that controls the hand making the movements (i.e., the one on the opposite
side). When first repeated, the activity in the cortex is at first reduced. This is due to
an effect called habituation, which is a common phenomenon in the nervous system.
However, when the finger sequence has been repeated a certain critical number of
times (in this case, a few dozen times), the signal in the cortex becomes larger, and
it stays larger during the training sessions on subsequent days. During this stage it is
thought that new connections are made between the nerve cells, resulting in a modifi-
cation of the motor map in the cortex (Kleim et al. 2004). Of course, the primary mo-
tor cortex is not the only part of the brain that is active during motor learning. Activity
is also seen in other areas such as the premotor and supplementary motor cortex, as
well as in the basal ganglia and the cerebellum (Hund-Georgiadis and von Cramon
1999). This occurs particularly during the early phases of motor skill activation and
declines as greater competence is achieved (fig. 8-3).
Interestingly, it has been shown that the activity in the cortex during the initial
stages of the finger-tapping task differs between pianists and nonmusicians. In pia-
nists there is much less activity in the supplementary motor and premotor areas, and
greater activity in the primary motor cortex (Haslinger et al. 2004). It is as if right
from the outset, the pianists are showing a pattern of activity that the nonmusicians
take some time to achieve, which presumably reflects their previous intensive training
in the control of fine finger movement. The supplementary motor cortex is thought
to be involved in the control of sequential movements performed in the absence of
visual feedback. It is more active when the task is complex; the reduced activity in
this region in the pianists therefore implies that the task is less demanding for them.
Comparisons of cortical activity between professional and amateur violinists playing
the same piece of music reveal a similar phenomenon (Lotze et al. 2003).
The learned movements used in instrumental performance are highly stereotyped.
The trajectories of points on the body surface can be accurately measured in two and
three dimensions using video analysis. Studies of violin players have revealed that the
size and timing of the trajectories of the fingers of the left hand (in a task requiring
minimal overall hand displacement) are almost identical when the same note sequence
is repeated several times. The same is true of the bow and the right arm in violinists
and cellists (Baader, Kazennikov, and Wiesendanger 2005; Turner-Stokes and Reid
1999; Wiesendanger, Baader, and Kazennikov 2006). Nevertheless, the synchroniza-
tion between the fingering and bowing is not absolutely precise. Even in expert players
 How the Performance of Music Affects the Brain 253

there is an average discrepancy of around 50 milliseconds between finger placement

and bow contact. However, there is a lag between the initial contact of the bow with
the string and the onset of the vibration (also typically around 50 milliseconds), and it
appears that the synchronization of the left and right hands is determined by the sound
that is heard and not by sensory feedback from the movements themselves. The size
of the acceptable error in synchrony therefore depends on how accurately the auditory
system can determine the beginning of the note (Baader, Kazennikov, and Wiesendan-
ger 2005). In the cello, left-hand excursions during playing can be of considerable mag-
nitude, and their accuracy has been investigated in players of various standards who
were asked to make movements of different speeds along the fingerboard in order to
play particular notes (Chen, Woollacott, and Pologe 2006). In studies of movement ac-
curacy, it is generally found that the greater the velocity of the movement and the force
used, the greater the error observed in the final position reached. Among the cellists,
however, neither the distance covered by the hand nor its velocity affected the average
accuracy of finger placement. This is probably an effect of the high level of training in
left-hand movement that cellists undergo. The accuracy of finger positioning was great-
est for the more proficient cellists but was also correlated with the accuracy of their
pitch perception. However, the variability in finger placement was not associated with
either the level of expertise or pitch perception, which suggests an underlying limita-
tion set by the motor system and its proprioceptive sensory feedback. Surprisingly, all
of the cellists in this study tended to play a little sharp, something that has also been
noted among wind players (Morrison 2000), but the reason for this is unknown.
Though the left side of the brain controls the right hand and vice versa, there is
some evidence that training of the nondominant hand may also cause activation of
cortical regions on the same side. This suggests that training one hand may improve
the performance of the other through communication between the two hemispheres
(Hund-Georgiadis and von Cramon 1999). Such transference has not been observed in
all studies, however, and remains controversial, but as we shall see later, occupational
focal dystonia in one hand can quickly appear in the other if it is used to carry out tasks
formerly assigned to the dystonic one. If, as has been proposed, this type of dystonia
is a maladaptive effect of overtraining, its transference to the other hand would be
consistent with a bilateral effect of training on the sensorimotor cortex.

Mental Rehearsal
For many professional musicians, mental rehearsal is an integral part of their prepara-
tion for performance. It is strongly promoted by some teachers and its use by famous
piano soloists such as Horowitz and Rubinstein is a matter of record. In neurobiological
terms, mental rehearsal and preparation can be seen to encompass several distinct ele-
ments. First, there is the interpretation of the score in terms of an internal representation
of sound (i.e., mentally hearing the music when reading the score). Second, there is the
committing of the score to memory, which includes not only the notation on the stave,
but also the marks of expression. The memory of the score can then be used to support
a virtual rehearsal of the movements required to perform it. Another form of mental
rehearsal is to read the score and mentally rehearse the performance movements. In its
254 Chapter Eight

most advanced stages, this may even be used to explore various options for expressive
interpretation. For this to be possible, the brain must create an internal image not only
of the movements but also of the precise effect they will have on the sound produced.
Though they are combined holistically, each of these tasks requires a different set of
mental skills, and we will first consider the mental rehearsal of playing movements. It
goes without saying that this needs a complete familiarity with and mastery of the instru-
ment and is therefore an option open only to advanced players (Lotze et al. 2003). This
form of physical imagery is not unique to musicians; it is widely used in sports that in-
volve complex stereotyped movements. High jumpers can often be seen mentally prac-
ticing before a crucial jump, and the complex and precision movements of divers also
benefit from this type of rehearsal. The effectiveness of mental practice in improving
aspects of the dynamics of movement, such as the accuracy of the movement trajectory,
has been verified experimentally (Yaguez et al. 1998). In one study, the effect of men-
tal rehearsal was compared with physical practice of the same simple musical passage
(Pascual-Leone 2001). The subjects practiced for two hours daily over five days, and
the effect on brain activation and the accuracy of performance was monitored. Though
the section of the primary motor cortex that drives the muscles of the hand remained
silent in the group carrying out only mental rehearsal, the size of the areas that would
have been controlling the muscles moving the fingers increased, while the threshold for
their activation was reduced. This was accompanied by a demonstrable improvement in
motor performance, though it was not as great as the one achieved by physical practice.
However, at the end of the study, a single session of physical practice in the mental
rehearsal group was sufficient for them to reach parity of performance with the physi-
cal rehearsal group. This result is in general agreement with those of other less physi-
ologically rigorous experiments that suggest that while mental practice is better than
no practice, it is not as good as actual practice (Gabrielsson 1999). However, mental
practice can be done in the absence of the instrument, and as part of an overall scheme
of rehearsal may, by reducing the amount of physical playing, potentially reduce the risk
of overuse injury (see chapter 3).
Clearly, one source of information missing in mental rehearsal is feedback from the
sensory endings in the fingers and their joints and muscles. The less experienced the
player, the more important this feedback will be; however, in experienced players in
whom the movements of the fingers have become automatic, its importance may be
reduced. There is also, of course, a lack of auditory feedback as there is no tangible
output from the virtual activity. The effect of its absence on performance accuracy has
been investigated in experienced pianists playing on a silent keyboard (Finney and
Palmer 2003). For substantial excerpts of previously learned pieces (e.g., music by
Bach, Beethoven, and Rachmaninov), there was no significant difference in error rate
between playing with or without sound. In simple sight-reading tests, however, though
the absence of auditory feedback had no effect on performance from the score, it did
have a deleterious effect on the accuracy of repeating the music from memory. For
those interested in developing and optimizing the skills of mental rehearsal, practical
information can be found in Connolly and Williamon (2004).
Outside the primary motor region, many of the same cortical areas that are normally
active during playing are also active to a lesser extent during virtual practice (fig. 8-
 How the Performance of Music Affects the Brain 255

3) (Lotze et al. 2003; Langheim et al. 2002; Meister et al. 2004). Functional imaging
studies have revealed activity in premotor and supplementary motor areas that are
thought to be involved in the generation of complex motor activity. The premotor area
is active when we silently “sing” a melody to ourselves, though not when the melody
is simply recalled. Perhaps surprisingly, there was no activity in the primary auditory
area during these experiments, despite the fact that virtual rehearsal generally requires
a vivid mental realization of the sound associated with the virtual “movements.” This
is in contrast to imagining a scene, which does produce activity in cortical visual ar-
eas. However, when the rehearsal involved real hand movements, activity was present
in the right primary auditory cortex and left auditory association cortex, even if the
hand was not in contact with the instrument and no sound was produced (Lotze et al.
2003). This suggests that some link exists between the primary motor cortex and the
auditory cortex (Bangert and Altenmuller 2003). The right primary auditory area is
the main region of the cortex involved in the perception of pitch, harmony, and tim-
bre, and the level of its activity during silent practice with actual finger movements is
greater in professional musicians than in the amateurs. A link in the opposite direction
between the primary auditory and primary motor cortices has also been demonstrated.
A study of advanced piano students demonstrated that listening to a piece of keyboard
music with which they were already familiar caused involuntary activity in the pri-
mary motor cortex even though the finger muscles were not activated (Haueisen and
Knosche 2001). Activity in the motor cortex occurred in the region controlling a finger
just before the note it would have played was sounded and so mirrored the activity that
would have been required for playing. No such response was seen in a control group
of similarly experienced singers who were not pianists. This type of connection would
undoubtedly support the ability to play music by ear.
As mentioned previously, one component of mental rehearsal is familiarization with
the score. A number of studies have investigated the relative effectiveness of different
score-learning strategies based either on playing the music through, examining the
score without playing it, or using a combination of both approaches (see Gabrielsson
1999). Unfortunately, differences in methodology make it difficult to draw general
conclusions from the results (Gabrielsson 1999). The studies had a variety of different
aims, and their design was not compatible with what is now known of the time scale
of and requirements for motor learning. Differences in the level of experience of the
subjects were also not taken into consideration. This is an important factor because
the way in which the score will be analyzed will depend on this (see below) as will
the degree to which certain note sequences are familiar to the individual through play-
ing technical studies and other repertoire. In addition, we have already gained some
inkling of the differences in brain development and activation patterns that may be
present as a result of different lifetime experiences of music.

Physical Practice
In the initial stages of learning a new piece, reading through the score may be used to
gain a clear idea of the notes and expression marks in the absence of the distraction
of the physical challenges of playing. This may take the form of studying the overall
256 Chapter Eight

structure of the piece and/or a careful examination of single phrases or their component
note sequences. The latter in particular should increase the probability of playing the
notes accurately at the first attempt and, in so doing, establishing the correct motor pro-
gram from the outset. If we relate this to what we already know of motor learning, this
should be of considerable advantage because the correct sequence of finger movements
will need to be repeated many times to generate the initial increase in cortical respon-
siveness that underlies the first stage of motor learning. If several variants are played
initially, this will at the very least slow the consolidation of the correct motor sequence
and at worst lead to the firm establishment of an incorrect variant that may persist for
some time as a learned alternative to the correct sequence (especially if the incorrect
sequence is easier to perform). The importance of the initial stages of learning is borne
out by observations of how high-level performers actually practice. In one study it was
found that different short segments of an eight-bar sequence were repeated more than
150 times by a pianist in the first practice session and 50 times in the second session
(e.g., fig. 8-5) (Chaffin and Lemieux 2004). The fact that it was not singled out for spe-
cial treatment on any subsequent session suggests that it had been effectively mastered,
demonstrating the efficacy of this highly focused approach to motor learning. In a
study of brass and woodwind students, it was found that structured, supervised practice
involving this type of approach was more effective than free practice where they were
left to their own devices and tended not to use such strategies (Barry 1992). One way of
ensuring accuracy right from the start is to play the notes slowly. It has, however, been
suggested that in keyboard instruments, which require complex bimanual coordination
of several voices, the motor activity used for slow playing is fundamentally different
from that used for fast playing (see Gabrielsson 1999). It has been proposed that in
such circumstances it would be necessary to adjust the technique of the slow playing
used when initially learning the music to compensate for this, or to develop early on the
facility to play short segments at close to final speed. However, this problem has yet to
be examined more objectively from a physiological standpoint.

Memorizing Music
Until the middle of the nineteenth century, it was unusual for soloists to perform
extensively from memory; the ability of musicians such as Franz Liszt and Clara
Schumann to do this was regarded as something of a phenomenon. Now most teach-
ers put considerable emphasis on the importance of learning the score by heart, and it
is regarded as unusual if a soloist plays directly from the music. From the viewpoint
of the biology of performance, freeing the brain from the requirement of reading the
score will not only eliminate a considerable cognitive load, but also allow the atten-
tion to be focused entirely on the sound being produced, which will undoubtedly be an
advantage in refining the performance. Of course, even when reading from the score,
many segments will already effectively have been committed to memory, as it is im-
possible to play fluently otherwise. In many cases there is simply not time to read all
the notes when playing at the correct tempo, and the written music is used simply as
a cue to bring forth memory of blocks of motor activity whose performance has been
refined through extensive practice.
 How the Performance of Music Affects the Brain 257

Learning a piece of music by heart involves several types of memory. We must learn
the written score, as well as retaining an auditory memory of how it sounds and a motor
memory of how it is played. Though the written score is the original source of infor-
mation, it is for the instrumentalist merely the first stage en route to constructing and
learning the motor program. As we have seen, the motor memory is constructed largely
by carrying out the activity repeatedly, but becoming familiar with the written score is
a more abstract task—the learning of a pattern—and is done in different regions of the
brain. At the most fundamental level, the motor memory is created from extensively
practiced note sequences based on elements of scales, arpeggios, and chordal progres-
sions (Halpern and Bower 1982). Grouping together a series of such operations that
are then remembered as a unit is called “chunking” and makes the memory task easier
as there are many fewer chunks to remember than notes. Similarly, learning the score
is more efficient if it is perceived to be composed of recognizable musical elements.
Music, like language, has a grammar based on scales and progressions, and within this
context many patterns are identifiable and predictable. Experienced players who are
familiar with musical syntax can use this to simplify the learning task, while less ex-
perienced players must create the motor pattern by reading the score by rote in a more
literal fashion (Aiello 2001). To use a linguistic analogy, it is easier to remember the
phrase “the dog was a golden retriever” than “golden the was retriever dog.” Further-
more, the anomaly in the phrase “the dig was a golden retriever” is immediately appar-
ent while less so in “golden the was retriever dig.” The memory processes that are used
to create a performance in the absence of the score can therefore operate at a variety of
levels. Some recent studies have investigated this through statistical analysis of practice
strategies, in one case of an individual professional pianist and in another, of groups of
pianists of different standards (Chaffin and Imreh 1997, 2002; Aiello 2001).
On the most basic level, a performance can be created by learning a piece of music
as a sequential series of chunks. This requires relatively little knowledge of the struc-
ture of the music and is often used by inexperienced musicians; however, it is a risky
strategy. Each new phrase is required to trigger the memory of the next and as the
sequence increases in length, it becomes progressively harder to do this. If there is a
break in the sequence due to a lapse in memory, the whole structure may fall. The only
recourse in such circumstances is to start the section again and risk another breakdown
at the same or a different point in the memory train, which may now be even more
likely due to the stress of the previous failure. However, if the structure of the music
is analyzed, it can be broken down into a series of elements, which are progressively
divided into smaller and smaller units according to a hierarchical scheme. The hier-
archy provides a context in which the “chunks” can be reassembled during playing
(fig. 8-5). As well as making it easier to memorize, this approach allows a musician
to recover from a memory lapse by rejoining the music at the beginning of the next
structural element. The bars at the beginning and end of each element become impor-
tant memory retrieval cues that require greater attention during learning than the bars
following. If musicians who have learned pieces in this way are shown isolated bars
from the piece, the cue bars are recognized more quickly and more accurately than
bars falling within segments, and this is reflected in the pattern of electrical activity
they evoke in the brain (Williamon and Egner 2004).
Figure 8-5
 How the Performance of Music Affects the Brain 259

Figure 8-5. The upper part of the figure shows the hierarchical structure of the Presto of Bach’s
Italian Concerto, illustrating the sequential way in which the memory of the piece is unpacked by
a pianist playing from memory. Particular attention needs to be given to transitional boundaries
between subsections and to the reappearance of similar, but not identical, musical elements (e.g.,
A1, A2 . . . A6). The lower part of the figure is a record of how a pianist went about preparing
and memorizing one section of the piece through intense “deliberate” practice. Each line indi-
cates a segment of music played continuously, and the practice session runs from bottom to top.
This session concentrated on a particular section of the movement (Cb); most playing episodes
start from its beginning. A sequence of episodes in which the playing is progressively extended
further into the section is repeated several times. This is interspersed with practice of short note
sequences that pose technical problems. As the session progresses, the sections played start to
concentrate on the transitions between adjacent segments and are designed to stitch them to-
gether to form a mental image of this part of the movement. Reproduced with permission from
Chaffin and Imreh (1997).

An insight into how this works in practice is provided in a study by Chaffin and
Imreh (1997, 2002), who recorded and analyzed rehearsal sessions during which a
piece of keyboard music (the Presto from Bach’s Italian Concerto) was learned by
a professional player (Imreh). The results are illuminated by the player’s subjective
comments on the process. One of the first tasks required in the playing of any piece
is to solve the motor problems posed by difficult note sequences, in this case the
fingerings. The music in question involves many sequences that, though basically
similar, are often subtly different. After some experimentation, the fingering solutions
ultimately used were not always the simplest, but the ones that could most consistently
be applied to similar phrases. In other words, the resolution of the motor difficulties
was viewed in the context of the higher-level problems of memorization. When the
fingerings had been converted to automatic motor memories, it was then possible to
turn the attention to discrete structural segments of the music, and then to link these
together into progressively longer sections. As we would expect from what has been
said previously, starts and stops during rehearsal occurred most frequently at cue bars
on the boundaries between musical segments.
Figure 8-5 shows the structure of the movement studied by Imreh. The initial theme
(A) is repeated in a number of guises throughout the piece, interspersed with other re-
peated sections (B, C, and D). On the one hand, this gives it a clear structure but on the
other, the final bar of theme A does not always signal a switch to theme B; what fol-
lows varies depending on where in the piece it occurs. This sets a particular challenge
for the memorization, which is reflected in the amount of time devoted to the practice
of key bars around these transitions (fig. 8-5). Imreh commented, “A lot of my later
practice . . . was practicing throwing those switches. My fingers played the notes just
fine. The practice I needed was in my head. As I approached a switching point I would
automatically think about where I was and which way to throw the switch.” (Chaffin
and Imreh 2002). The analogy of “throwing a switch” underlines the fact that motor
sequences of the chunks had by this stage become fully automatic.
The emergence of the types of memory skills seen in expert musicians can be ob-
served by examining patterns of practice and levels of achievement seen in individuals
at different stages of musical training. Williamon and Valentine (2002) compared pia-
nists grouped into four levels of experience at an early stage in their musical careers,
260 Chapter Eight

graded according to the scheme of the Associate Board of the Royal School of Music.
These represented grades 1–2 (level 1), grades 3–4 (level 2), grades 5–6 (level 3), and
grades 7–8 (level 4). Each group was asked to learn and memorize a piece of music
of appropriate difficulty. All had some awareness of how the structure of the pieces
could be divided into segments, either through guidance from teachers or through their
own objective or intuitive understanding. In all groups but level 1, starts and stops
during practice sessions tended to coincide with structural boundaries in the music as
the familiarity with it increased, but this was particularly marked at the higher levels
of proficiency. Starts and stops at difficult bars declined in all groups but level 1 with
increasing practice, presumably reflecting the progressive integration of these sections
into the rest of the piece as the technical problems were overcome. It therefore seems
that the same hierarchical memory retrieval systems seen in expert performers are be-
ing employed, in at least a basic form, by all but the least experienced players. From
a pedagogical point of view, it is apparent that an ability to understand the underlying
musical structure of a piece develops naturally with experience. These results also
suggest that the development of these memory retrieval strategies can be encouraged
even at an early stage of musical training, which in view of the complexity of the
hierarchical schemata required for the memorizing of typical recital pieces may have
considerable long-term advantages for the student.
High levels of practical musical expertise are not easily won. By the age of twenty,
a proficient pianist is likely to have accumulated in the order of ten thousand hours of
practice, and even achieving a standard equivalent to grade 8 typically requires around
a third of this (Sloboda et al. 1996), at least for a first instrument. However, it is not
the time spent in practice alone that is significant, but also its quality. For maximum
benefit the practice must also be “deliberate,” that is to say, rather than simply playing
through the music, it must be highly structured with the attention focused on improv-
ing a particular aspect of performance. We have already encountered this notion in
our consideration of motor learning, in which attention is a key factor (Noppeney et
al. 1999), as well as in the advantages of supervised practice (Barry 1992). Need-
less to say, this requires considerable effort and mental concentration, and in view
of the fact that the benefits may take some time to become evident, the demands of
the practice sessions must be sustainable over a prolonged period from both a mental
and a motivational point of view. Consequently, sessions of deliberate practice tend
to last only between one hour and ninety minutes. In one study, the amount of ef-
fort required appeared to be reflected in the greater incidence of daytime napping in
elite musicians. They also tended to play more in the morning when presumably they
were most fresh (Ericsson, Krampe, and Tesch-Romer 1993). In addition, the highest
achievers not only concentrated more on deliberate practice but were more consistent
in their practice schedules.
Comparisons between players of different standards have consistently revealed that
the more experienced practice in longer sessions. In many studies, of course, age tends
to increase in parallel with expertise, but the increase in session length does not ap-
pear to be simply a consequence of greater patience and determination. The improved
physical stamina that age brings may be a factor; however, it may also be that once the
more mundane problems of basic technique have been mastered, the challenges that
 How the Performance of Music Affects the Brain 261

open up in terms of interpretation and memorization provide many different focuses

of attention, making practice more varied and interesting (Williamon and Valentine
2000).

OCCUPATIONAL FOCAL DYSTONIA

Established instrumental musicians in midcareer sometimes develop a painless but

unaccountable inability to control the hand. This often starts as a feeling of heaviness
in the fingers but becomes progressively more debilitating so that ultimately the player
may only have to pick up or sit at the instrument to find the fingers involuntarily tak-
ing up a hyperflexed or hyperextended posture that makes playing impossible. These
are the symptoms of occupational focal dystonia.
Dystonia refers to abnormal involuntary and uncontrollable muscle contractions
that are most often seen in the hand or in muscles of the neck or face. Though it is
usually painless initially, the extreme contraction that may ultimately develop in sets
of antagonistic muscles can cause considerable discomfort. The excessive and inap-
propriate muscle activity may generate abnormal postures of the upper limb or body, a
twisting of joints such as the wrist or neck, rhythmic movements, or tremor. Dystonia
is said to be focal when it affects only a single muscle or a small group of muscles,
and this is generally task specific, appearing particularly during skilled movements
that have been practiced habitually over a long period of time, for example, as part of
one’s occupation. In musicians, though it may be triggered just by holding or touching
the instrument, it may not affect a similar movement used to pick up and use a tool or
a piece of cutlery. The symptoms tend to come on gradually and in the initial stages
may be apparent to the player only as a subjective feeling of difficulty, a heaviness in
the arm, or a resistance to movement in the fingers. At first the consequences may not
be obvious to the listener, and the condition sometimes takes years to develop fully;
in other cases the dystonia may become debilitating in a matter of months (Tubiana
2000).
Occupational focal dystonias have a long history, and a brief but informative ac-
count of this can be found in Hochberg, Harris, and Blattert (1990). The first probable
description of focal dystonia in Western medicine was written in the early eighteenth
century by the father of occupational health studies, Bernardini Ramazzini. He de-
scribes an intense fatigue and a loss of power in the right hand of a notary brought
on by writing. When the notary tried to solve the problem by using the other hand, it
soon became affected as well. This type of dystonia was well known in the nineteenth
century and was given the name scrivener’s palsy (widely known as writer’s cramp).
An account by Solly in 1864 as quoted by Hochberg, Harris, and Blattert (1990) de-
scribes its essential features: “It comes on insidiously, the first indication often being
only a painful feeling in the thumb or forefinger of the writing hand accompanied by
some stiffness. . . . The loss of power is not sudden as in a paralytic stroke; nor is it a
complete paralysis of any group of muscles. The paralyzed scrivener, though he can-
not write, can amuse himself in his garden, can shoot, and cut his meat like a Christian
at the dinner table; indeed he can do anything he likes, except earn his daily bread
262 Chapter Eight

as a scribbler.” This was only one of many work-related dystonias (or occupational
neuroses as they were sometimes known) that were described at this time. These cov-
ered more than fifty occupations, but musicians were said to be second only to writers
in their vulnerability to it (Fry 1986). Its origins have long remained obscure, but in
recent years the study of affected musicians with modern brain scanning techniques
has done much to increase our understanding of focal dystonia.
Because focal dystonia has been widely discussed in recent literature on musicians’
health, instrumentalists with an interest in arts medicine may have an exaggerated idea
of how common it is. The incidence can be difficult to determine from reports from
well-known clinics specializing in musicians’ health problems because these attract
difficult cases that have not received effective diagnosis or treatment elsewhere. As
dystonia is frequently misdiagnosed in the first place, its sufferers are likely to be
overrepresented among their patients. Therefore, though their reports would appear
to suggest that 5 to 15 percent of musicians with motor problems have focal dysto-
nia, the figure is undoubtedly much lower. The best evidence of its incidence among
professional musicians suggests that this is in the order of 1 in 200 to 1 in 500 (Lim,
Altenmuller, and Bradshaw 2001), which is around ten times that seen in the general
population (Pullman and Hristova 2005; Wynn Parry and Tubiana 1998). Several
studies have collated information on the characteristics of focal dystonia among mu-
sicians, and a comprehensive review of the more important reports can be found in
Lim, Altenmuller, and Bradshaw (2001), who also assess current theories of its origin.
However, for a shorter and more accessible overview, readers are directed to Jabusch
and Altenmuller (2006) which can be freely downloaded from the journal website
http://www.ac-psych.org/. Focal dystonia most commonly appears in the fourth de-
cade of life (Lim, Altenmuller, and Bradshaw 2001; Hochberg and Hochberg 2000;
Wynn Parry and Tubiana 1998), though it is certainly not unknown at an earlier age
(Jankovic and Shale 1989). This means that it generally affects musicians in midca-
reer, when they already have a sound and long-established technique. In contrast to
their younger colleagues, they are unlikely to be pushing hard to develop a profes-
sional level of proficiency or to be competing vigorously to gain their first foothold
in the musical world. There is a distinct but unexplained gender bias in the expression
of dystonia. Different studies suggest that men are two to six times more likely to be
affected (Jabusch and Altenmuller 2006). Most of those affected have problems with
coordination of the hand, and there is a high incidence of dystonia among pianists and
guitarists, though the hands of other string players and of woodwind and brass players
can also be affected. In addition, focal dystonia of the embouchure (see chapter 6) is
not uncommon in wind players (Frucht et al. 2001).

Symptoms of Focal Dystonia in Musicians

The onset of dystonia is usually gradual and painless. Among keyboard players there
is initially a loss of facility in passages that may have been executed without prob-
lems for decades (Hochberg and Hochberg 2000; Tubiana 2000). In rapid passages,
particularly those containing arpeggios, some of the fingers may appear to lag behind
the others in an unexplainable way. Problems may also be experienced with playing
 How the Performance of Music Affects the Brain 263

octaves. In string players, difficulties may be found with double stopping as well as
passage work, in which a finger or fingers become slow to lift. As the condition pro-
gresses, the affected fingers often assume a more flexed posture so that the edge of the
nail strikes the piano key or it becomes difficult to raise the finger from between the
strings. Finally, the fingers may become completely curled under the palm so that the
outer surface of the fingernail instead of the pad strikes the key (Candia et al. 2002). In
other cases, the recalcitrant fingers instead become involuntarily extended and cannot
be brought down to touch the key or string. The later stages of dystonia are sometimes
characterized by a fierce simultaneous contraction of antagonistic flexor and extensor
muscles of the digits, which causes fatigue and pain. Interestingly, in patients who play
more than one instrument, the dystonia may affect the playing of only one. Activities
such as typing, which one might think would share many similarities with keyboard
playing, may also be unaffected (Wynn Parry and Tubiana 1998). The typical site of
the dystonia varies depending on the instrument. Wynn Parry and Tubiana report that
in pianists, it is the right hand that is most usually dystonic (70 percent), in which case
the most commonly affected fingers are those closest to the little finger. When the left
hand is involved, the affected digits tend to be on the other side. Among guitarists, it
is the index and ring fingers of the right hand that are most likely to become dystonic,
while in violinists, three-quarters of cases involve the left hand.

Theories of the Origin of Dystonia

Occupational focal dystonias are often described by the medical profession as id-
iopathic, meaning of unknown cause. This should not make you immediately lose
confidence in your doctor! Many well-known illnesses are idiopathic, and though
the science of medicine has progressed greatly over the last century and a half, so
has our recognition of the incredible complexity of the physiology of the body and
its interactions with its environment. Though the origin of focal dystonia is unclear,
our knowledge of its symptoms goes much deeper than the abnormal movements it
causes. There are many clues as to what may lie at the root of the problem, but none
has yet led to a definitive conclusion. Some researchers have proposed a strong causal
link between focal dystonia and repetitive strain injury, and one group has even gone
as far as to define a condition known as “repetitive strain–induced focal dystonia”
(Byl and McKenzie 2000). This theory is not currently widely accepted for human
focal dystonia, as the relationship between dystonia and injury, regardless of how it is
caused, is for the most part unclear. While some anecdotal evidence has been put for-
ward to support the idea that a variety of injuries might trigger the appearance of dys-
tonia (Jankovic and Shale 1989), a lack of objectivity in the way that much of the data
have been collected, and the absence of any rationale to explain the theory, seriously
undermines its credibility. However, there is quite good evidence that nerve compres-
sion, particularly when it involves the ulnar nerve, is a risk factor (Charness, Ross,
and Shefner 1996). One factor that makes this theory convincing is that in dystonic
musicians with demonstrable ulnar nerve pathology, the affected muscles are particu-
larly those controlling the little and ring fingers. The ulnar nerve supplies the small
muscles within the hand that act on these fingers as well as the skin that covers them.
264 Chapter Eight

Furthermore, treatment to relieve ulnar nerve compression can markedly improve or

even eliminate the dystonia, though this may take many months. Unfortunately, in
other forms of dystonia of the hand in which the affected muscles are supplied by the
radial or median nerves, the same link with nerve compression has not been found.
The observation that dystonic symptoms can be generated in monkeys by inducing
them to carry out repetitive, stereotyped hand movements appears highly significant
(Byl and Melnick 1997), as it is precisely this type of activity that underlies musical
practice and performance. However, the context dependency of the symptoms remains
a mystery. Dystonia sometimes appears in musicians following a marked increase in
the intensity of practice, a substantial change in technique, or upon switching to a dif-
ferent model of instrument that may make slightly different physical demands than the
one used previously. Changes in the physical sensations that these new circumstances
generate may also be important. In such circumstances it cannot be ruled out that
there has been some unidentified trauma, related perhaps to the change in technique
or overintensive practice, but this would be inconsistent with the observation that
the onset of dystonia is typically painless. In addition, the average age of onset for
overuse injuries among musicians is in the midtwenties, while for dystonia it is in the
mid- to late thirties (Tubiana 2000). It is also thought that an obsessive perfection-
ism or anxiety about performance may be a predisposing factor for the emergence of
dystonia (Jabusch and Altenmuller 2006), though this needs to be more rigorously
investigated. This may lead the player to carry out repeated but futile attempts to ex-
ceed the biomechanical limits of the hand, rather than to explore alternative solutions
to problems of fingering.
There is a growing acceptance that focal dystonia is not fundamentally a condition
of muscle or peripheral nerve malfunction, but is primarily related to changes in the
parts of the brain that control and coordinate movement. We have already seen that
the wiring of the brain is not fixed, but can be shaped by experience and that this
plasticity can alter the sensory and motor maps within the cortex. Studies of the brain
of monkeys and humans with focal dystonia reveal changes in the organization of the
area of the primary sensory cortex receiving information from the affected region. In
monkeys that develop dystonia after being trained to carry out repetitive hand move-
ments, there is a degradation in the sensory map of the hand. The near-simultaneous
stimulation of adjacent fingers during tasks that require close attention to maintain
accuracy appears to be the most significant factor underlying these changes. The re-
ceptive fields of the nerve cells that map the fingers onto the cortex can be enlarged by
a factor of ten to twenty, resulting in considerable overlap between the representation
of adjacent fingers and even between the front and back of the hand (Byl and Melnick
1997). Receptive fields may also grow to cover several finger joints. The end result
is that the brain can no longer tell what part of the finger, or even which finger, is
being touched. Investigations of musicians and nonmusicians with focal dystonia of
the hand reveal similar maladaptive changes in the cortical mapping. The finger repre-
sentations of the dystonic hand are much closer together and either overlap or appear
in random order (Bara-Jimenez et al. 1998; Elbert et al. 1998; Byl, McKenzie, and
Nagarajan 2000; Butterworth et al. 2003). A study by Elbert et al. (1998) has also sug-
gested that there is some degradation in the sensory maps for the nonaffected hand of
 How the Performance of Music Affects the Brain 265

dystonic patients, though this is not as marked as that of the affected hand. The notion
that the sensory maps in each hemisphere are not entirely independent is supported
by observations on readers of Braille, who use several fingers simultaneously to read
text. The changes this produces in the cortical representation of these fingers is often
reflected in the representation of the other hand, even though it is not used for Braille
(Sterr et al. 1998). This may partly explain why dystonia sometimes appears quite rap-
idly in the previously nonaffected hand when dystonic patients try to use it for tasks
previously carried out by the dystonic hand (Lim, Altenmuller, and Bradshaw 2001).
In brass players with dystonic embouchures, a distortion has also been reported in the
cortical mapping of the lips, with a decrease in the ability to detect gaps between fine
gratings applied to the upper, though not the lower, lip (Hirata et al. 2004).
Focal dystonia is of course a disorder of motor control, so if the idea that it is
caused by changes in the sensory map is valid, it should theoretically be possible to
demonstrate some reconfiguration of the motor cortex. In the normal brain, large-scale
changes in the motor map have been demonstrated experimentally, and it is clear that
these can take place very rapidly, in fact, over a period of just a few hours (Kandel,
Schwartz, and Jessell 2000). In dystonic musicians, however, it has not been possible
to detect changes in the motor map of the hand. The reason for this is probably that,
as we have already seen, there is not the same precise topographical order in the motor
map that exists in its sensory partner. Though there is a region devoted to the hand,
the representations of muscles moving individual fingers are not discrete, but are nor-
mally intermingled. Contraction of a single muscle is reflected in activity at a number
of sites in the hand area of the motor cortex and so any changes wrought by dystonia
would be difficult to detect.
Changes in other motor areas of the brain such as the premotor cortex and in the
basal ganglia have also been shown in dystonic patients, including musicians (Lim,
Altenmuller, and Bradshaw 2001; Lim et al. 2004; Ibanez et al. 1999). For example,
during the execution of tasks that induce dystonic symptoms in musicians, activity
in the premotor cortex fell below normal levels while that in the primary motor and
sensory cortices was increased (Pujol et al. 2000). A similar result has been reported
in dystonic nonmusicians (Lim, Altenmuller, and Bradshaw 2001). There are several
other types of dystonia, most of which are thought to be caused by impaired function
of the basal ganglia. This may be the result of trauma such as direct physical injury
or damage due to stroke (loss of blood supply), or to changes in neurotransmitter lev-
els. One component of the basal ganglia known as the putamen contains a body map
incorporating both sensory and motor information (Romanelli et al. 2005). A study
of one group of patients with focal dystonia of the hand (in this case writer’s cramp)
clearly demonstrated that this map had become disorganized on the side of the brain
controlling the dystonic limb, but not on the other side (Delmaire et al. 2005).
One consequence of the intensive training needed to perfect the rapid and continu-
ous finger movements required by instrumentalists is a reduction in the strength of
inhibitory pathways within the motor cortex. In nonmusicians, sensation originating
from one muscle in the hand reduces the excitability of cortical areas controlling other
hand muscles. The training of hand movements that musicians undergo appears to
alter this so that muscles that need to be active simultaneously during playing do not
266 Chapter Eight

inhibit each other so strongly. In dystonic musicians, this loss of inhibition progresses
even further to include muscles that are not usually active together. Thus, in the dys-
tonic musician, a process that may facilitate hand control appears to have progressed
too far, resulting in a maladaptive state (Rosenkranz et al. 2005). Patients with writer’s
cramp, a form of dystonia that has some similarities with musicians’ hand dystonia, do
not show this change. This may reflect the fact that most writer’s cramp patients do
not have a history of excessive writing; in other words, the origin of their condition,
though still unknown, is not due to overtraining, though it may be linked to overuse.
A similar loss of inhibition is found in the interaction between the hand control
areas in the opposite hemispheres of the brain. Activity in the primary motor cortex
on one side of the brain normally has an inhibitory effect on the equivalent areas on
the opposite side; however, in instrumental musicians the level of this inhibition is
reduced (Nordstrom and Butler 2002). The reason it is reduced in normal musicians
may be the necessity of coordinating very precisely the action of the fingers on each
hand. A similar reduction in inhibition is also seen in dystonic nonmusicians. This
again suggests that a training-induced change in brain circuitry, while advantageous
in moderation, may contribute to the maladaptive changes that underlie dystonia if
unbridled. It may also be another reason why dystonia can develop rapidly in the non-
affected hand when it takes over tasks previously performed by the dystonic one. Our
best guess about the origin of musicians’ focal dystonia is therefore that it is a disorder
of the integration of sensory and motor activity underlying musical performance. This
emerging perception has given new impetus to attempts to devise effective treatments
for the condition.

Predisposing Factors
In view of the uncertainty concerning the origins of focal dystonia, theories concern-
ing possible predisposing factors need to be approached with some caution. There
is excellent evidence that some families carry a gene that strongly predisposes them
to develop a form of early-onset dystonia, which typically affects teenagers (Lim,
Altenmuller, and Bradshaw 2001); however, whether focal dystonia has a significant
genetic component has not been studied in detail. One report suggests that a minority
of dystonic musicians (perhaps around 10 percent) may have close family members
with various forms of focal dystonia (Schmidt et al. 2006). Not all of the affected rela-
tives were musicians, and the condition most commonly manifested itself as writer’s
cramp. Interestingly, in two cases the relatives were professional pianists who suffered
from writer’s cramp but had no playing-related symptoms.
There has been some speculation about how psychological factors such as stress or
anxiety may contribute either directly or indirectly to the development of focal dysto-
nia. It has been proposed that the resistance of focal dystonias to therapy is related to
the integration of the motor programs with the emotional and memory-forming ele-
ments of the limbic system (see chapter 10). A link between the physical aspects of
playing and emotional responses could be a potent force for fixing motor memories,
and might support a connection with performance-related stress (Lim, Altenmuller,
and Bradshaw 2001).
 How the Performance of Music Affects the Brain 267

The repetitive nature of the hand movements required in instrumental perfor-

mance, especially when repeated for hours at a time, may lend itself to the genera-
tion of dystonia in some individuals. The risk factors listed by some authors read like
a standard catalogue of poor biomechanical technique: postural tension, instability
in the shoulders and other joints, working at the extreme ends of the range of joint
motion, heavy use of alternating flexor and extensor muscle activity, and playing
with nerve compression syndromes (Byl and Melnick 1997). An investigation of the
structure and joint movement ranges of the hands of musicians with focal dystonia
revealed that many may be pushing the biomechanical limits of what their hands
are capable of (Wilson, Wagner, and Homberg 1993); however, in the absence of
control data from a large population of unaffected musicians, this remains only
suggestive and not all of those working in the field of focal dystonia believe that
it is related to what in the broadest terms might be called faulty technique (Alten-
muller 2003). Nevertheless, it underlines the importance for teachers of taking into
consideration the physical limitations of their pupils when trying to optimize their
technique. The link reported between ulnar neuropathy and focal dystonia might be
explained by the related change in sensory input coupled with the compensatory
movements used to overcome the ensuing muscle weakness, which together may
disrupt the well-established motor programs controlling highly trained stereotyped
movements. However, one is left wondering why there is no obvious link with other
nerve compressions.

Treatments
Until recently there has been no treatment for focal dystonia that significantly restores
the ability to play the instrument at a high level. A small number of recent reports do
describe a possible way forward, but their development is at an early stage and they
have been so far tried on only a small number of individuals. One must therefore
remain cautious about how great an impact they will ultimately make on treatment
and rehabilitation. We will first consider the traditional approaches to the treatment of
occupational focal dystonia.
Treatment of any medical condition is dependent on three factors: 1) an accurate
diagnosis, 2) understanding the cause, and 3) having a procedure or a drug that is
effective in correcting the problem. In the past, treatments for occupational focal dys-
tonia have had limited effectiveness because of difficulties with each of these three
factors. The first step is, of course, an accurate diagnosis. In one survey of musicians
with focal dystonia, several had undergone what appeared to have been inappropriate
surgery (Hochberg, Harris, and Blattert 1990). Among another group of 189 patients,
40 percent had been wrongly diagnosed initially (Altenmuller 2003). Altenmuller also
stresses that practitioners working with dystonic musicians should discuss the possible
treatments realistically with their patients without raising false hopes of rapid recovery
and should suggest a pragmatic approach to treatment based on a number of possible
therapies. These should take into account the particular situation of the individual and
the physical requirements of his or her instrument. Given the possible importance
of psychological factors, emotional support is important and the patient should be
268 Chapter Eight

encouraged to find some way of coming to terms with the limitations imposed by the
dystonia and discouraged from indulging in useless therapies.
The term “dystonia” refers to a set of symptoms and not to a single condition, so
the several types of dystonia (of which occupational focal dystonia is but one) do not
necessarily have the same underlying cause, though they may share some common fea-
tures. A number of drug treatments are used for dystonias, but most have been devel-
oped for other conditions. Finding one that has some beneficial effect remains largely
a question of trial and error. Furthermore, the effectiveness of any drug in a particular
patient may decline over time. Many of the drugs used are powerful ones that act by
interfering with different classes of neurotransmitters (the chemicals used by neurons
to communicate with each other; see chapter 7). Unfortunately, these transmitters have
many different roles in the nervous system other than the control of movement, so
drug treatments often have significant side effects. Furthermore, some may have only
a palliative effect on the condition, and while they may reduce symptoms such as the
discomfort of chronic muscle contractions, they do not provide a cure (Hochberg, Har-
ris, and Blattert 1990). The major classes of drugs used are as follows:

1. Benzodiazepines and drugs such as baclofen. These act on receptors for the inhibi-
tory neurotransmitter GABA to reduce nerve cell activity within the nervous sys-
tem. They can be used to induce muscle relaxation by making motor neurons less
active. The major side effect of the benzodiazepines is drowsiness.
2. Anticholinergic drugs, which block the neurotransmitter used by the motor neu-
rons. This neurotransmitter is also found in many brain neurons, and it is these
that are targeted by the drugs used to treat dystonia. The drug usually employed is
trihexyphenidyl (Artane). The rationale for its use is that it damps down circuits
in the brain that become overactive when levels of the transmitter dopamine (see
below) are reduced. Whether dopamine levels do change significantly in focal
dystonia is unclear, but Artane does relieve symptoms in some patients, though it
cannot be tolerated for long periods. Side effects can include confusion, hallucina-
tions, drowsiness, memory loss, and dry mouth, among others.
3. Drugs that mimic the effects of dopamine, a neurotransmitter involved in nerve
cells within the motor pathways of the central nervous system (e.g., their loss is the
cause of Parkinson’s disease). Some forms of dystonia can be controlled by these
drugs, though side effects may include slowing of movements and depression.
They are usually less effective than anticholinergics for focal dystonia.

The chronic muscle contraction caused by dystonias can also be treated by the injec-
tion of botulinum toxin directly into the muscles affected. This drug is probably most
familiar to the general public under the name of Botox, which is used cosmetically
to paralyze muscles that wrinkle the skin of the face. The dosage must be carefully
controlled so that it only weakens rather than paralyzes the affected muscles; however,
the problem for dystonic musicians is that it is difficult to regulate this sufficiently to
preserve the high degree of control required for playing. It is most effective when used
on the small muscles that lie within the hand, as there is less chance of affecting the
movement of adjacent fingers than when the injections are made into the large mus-
 How the Performance of Music Affects the Brain 269

cles of the forearm. However, many of the intrinsic muscles of the hand are involved
in the rapid lateral movements of the fingers, which are used constantly by woodwind,
string, and keyboard players. For woodwind players, for example, this is particularly
crucial where the little fingers must depress several keys, but it is also important for
the thumb and the other fingers. Therefore, for these players, Botox injection into hand
muscles may be inappropriate (Altenmuller 2003). Finally, even where Botox is used
successfully, patients may ultimately develop antibodies that will block its effects.
The discovery of the changes in cortical mapping that are associated with occupa-
tional focal dystonia has led to the development of a variety of new treatments based
not on drugs but on physical retraining aimed at reversing these trends. The most
puzzling aspect of dystonia is that although it is possible to modify sensory and motor
maps in the brain very rapidly, it is extremely difficult to reverse the changes associ-
ated with focal dystonia. This problem has remained stubbornly resistant to investiga-
tion. At the moment there is no consensus on the optimum approach to take; it is also
not clear what the limits to recovery will be. The techniques tried so far are based on
sensory reeducation of the hand, either alone or together with motor exercises. The
tasks used for sensory retraining are designed to require mental attentiveness, as this
appears important for the remolding of sensory maps in the cortex. One approach
uses Braille reading for sensory retraining without any motor training (Zeuner et al.
2002). The subjects of this study were nonmusicians who exhibited some form of
writer’s cramp. They had an initial training period of eight weeks, after which some
continued with daily practice for up to six months. Spatial acuity (the ability of the
fingers to distinguish closely positioned bars) improved significantly, suggesting that
some reconfiguration of the sensory map was achieved. Those who persevered with
the regimen showed considerable improvement in writing, but those who stopped
reverted quite quickly to their dystonic state. In a second study, subjects who suffered
from a variety of occupational dystonias were trained using a much heavier schedule
involving a range of sensory discriminative tasks that required the tactile identification
of objects and patterns by the hands and fingers. This was accompanied by a program
aimed at reducing the aberrant motor activity of dystonia and improving general fit-
ness and posture (Byl and McKenzie 2000). Improvements in sensory discrimination
and motor accuracy were again evident, and though movements remained slower than
normal, most of the subjects were able to return to work. However, none required the
high degree of control needed by professional musicians.
There has been only a single study directed at applying retraining techniques to
musicians, who are probably one of the most challenging groups in terms of rehabili-
tation. It used an approach called sensory-motor retuning (SMR) (Candia et al. 2002,
2003; Taub, Uswatte, and Elbert 2002). This study involved six pianists, two guitar-
ists, two flutists, and an oboist, all suffering from dystonia of the hand. In each case,
the main finger or fingers that were being used to compensate for the dystonic one
were identified and held immobile with splints, leaving the dystonic finger to cope
on its own in a series of highly structured exercises involving sequential movements
of this and the other free fingers at a series of different tempos. After these exercises,
the subjects played first simple and then more complex pieces with the hand uncon-
strained, during which they generally noticed considerable relief from their symptoms.
270 Chapter Eight

Initially the periods of free practice were short, lasting only fifteen to thirty seconds,
but were gradually increased to fifteen minutes. It should be noted that SMR differs
from constraint-induced movement therapy (CIMT), which is sometimes used for the
rehabilitation of stroke victims. In CIMT the compensating part of the body is im-
mobilized for prolonged periods so that the patient is forced to use the affected part
for all tasks. In SMR, by contrast, the exercises carried out during immobilization are
structured, with the aim of reeducating the affected joints to maintain nondystonic an-
gles, while between training sessions, free-playing sessions are encouraged so that the
newly reinforced finger postures are incorporated into the context of normal playing.
During a period of up to a year of sensory motor retraining, the dystonic finger
was required to perform repetitive movements demanding coordination of different
muscles for an hour or more each day. Many of the subjects showed signs of im-
provement even after an initial eight days of intensive training, though maintaining
and extending this further required that the exercises be continued regularly for many
months. There was a significant improvement among the pianists and guitarists who
persisted with the training, and some reached a level close to normality. Subjectively,
these participants felt that the results were a clear improvement on progress they had
made with anticholinergic drugs or Botox injections, and of course there were no
side effects from retraining. Objective analysis also demonstrated normalized finger
movement. There were also indications of a normalization of the sensory mapping of
the fingers within the brain (Candia et al. 2003). Unfortunately, for reasons that are
unclear, the therapy produced no improvement in wind players with hand dystonia,
and there is no obvious way that such an approach might be extended to embouchure
dystonia; nevertheless, SMR therapy is a promising treatment that warrants further
trials involving much larger numbers of affected musicians.
Another approach to reducing dystonic hand movements in musicians is limb im-
mobilization (Priori et al. 2001; Pesenti, Barbieri, and Priori 2004), in which a splint
is used to completely prevent any movement of the wrist and hand on the affected
side for a period of four to five weeks. Because immobilization leads to a reduction
in the cortical representation of the affected limb, the rationale behind the treatment
is that it will reduce the overlap between the fingers in the cortical map of the hand.
Such a long period of immobilization leads to considerable loss of muscle strength,
which takes several weeks to be restored, but the treatment was reported to produce
considerable long-term improvement in four of seven musicians who took part. It ap-
peared to be most effective in those whose dystonia was of most recent onset. How-
ever, the treatment needs a more detailed evaluation, and so far no cortical mapping
studies have been done to confirm that the desired changes are actually taking place
as a result of treatment.
In conclusion, it can be seen that new ideas about the origin of focal hand dystonia
have led to the exploration of new therapies based on a rational understanding of the
condition. Though there is some way to go before the causes of this form of dystonia
are fully understood, and though further work will be needed to establish the best
strategies for rehabilitation, the possibility that effective treatments will ultimately be
developed to replace the largely palliative therapies that are currently available looks
increasingly hopeful.
 How the Performance of Music Affects the Brain 271

REFERENCES

Aiello, R. 2001. Playing the piano by heart. From behavior to cognition. Ann NY Acad Sci
930:389–93.
Altenmuller, E. 2003. Focal dystonia: Advances in brain imaging and understanding of fine
motor control in musicians. Hand Clin 19 (3):xi, 523–38.
Amunts, K., G. Schlaug, L. Jancke, H. Steinmetz, A. Schleicher, A. Dabringhaus, and K. Zilles.
1997. Motor cortex and hand motor skills: Structural compliance in the human brain. Hum
Brain Mapp 5:206–15.
Amunts, K., G. Schlaug, A. Schleicher, H. Steinmetz, A. Dabringhaus, P. E. Roland, and
K. Zilles. 1996. Asymmetry in the human motor cortex and handedness. Neuroimage 4 (3
Pt 1):216–22.
Baader, A. P., O. Kazennikov, and M. Wiesendanger. 2005. Coordination of bowing and fin-
gering in violin playing. Brain Res Cogn Brain Res 23 (2–3):436–43.
Bangert, M., and E. O. Altenmuller. 2003. Mapping perception to action in piano practice: A
longitudinal DC-EEG study. BMC Neurosci 4 (1):26.
Bangert, M., and G. Schlaug. 2006. Specialization of the specialized in features of external
human brain morphology. Eur J Neurosci 24 (6):1832–34.
Bara-Jimenez, W., M. J. Catalan, M. Hallett, and C. Gerloff. 1998. Abnormal somatosensory
homunculus in dystonia of the hand. Ann Neurol 44 (5):828–31.
Barry, N. H. 1992. The effects of practise strategies, individual differences in cognitive style,
and gender upon technical accuracy and musicality of student instrumental performance.
Psychol Music 20:112–23.
Bengtsson, S. L., Z. Nagy, S. Skare, L. Forsman, H. Forssberg, and F. Ullen. 2005. Extensive
piano practicing has regionally specific effects on white matter development. Nat Neurosci
8 (9):1148–50.
Butterworth, S., S. Francis, E. Kelly, F. McGlone, R. Bowtell, and G. V. Sawle. 2003. Abnor-
mal cortical sensory activation in dystonia: An fMRI study. Mov Disord 18 (6):673–82.
Byl, N. N., and A. McKenzie. 2000. Treatment effectiveness for patients with a history of re-
petitive hand use and focal hand dystonia: A planned, prospective follow-up study. J Hand
Ther 13 (4):289–301.
Byl, N. N., A. McKenzie, and S. S. Nagarajan. 2000. Differences in somatosensory hand or-
ganization in a healthy flutist and a flutist with focal hand dystonia: A case report. J Hand
Ther 13 (4):302–9.
Byl, N. N., and M. Melnick. 1997. The neural consequences of repetition: Clinical implications
of a learning hypothesis. J Hand Ther 10 (2):160–74.
Candia, V., T. Schafer, E. Taub, H. Rau, E. Altenmuller, B. Rockstroh, and T. Elbert. 2002.
Sensory motor retuning: A behavioral treatment for focal hand dystonia of pianists and gui-
tarists. Arch Phys Med Rehabil 83 (10):1342–48.
Candia, V., C. Wienbruch, T. Elbert, B. Rockstroh, and W. Ray. 2003. Effective behavioral
treatment of focal hand dystonia in musicians alters somatosensory cortical organization.
Proc Natl Acad Sci USA 100 (13):7942–46.
Chaffin, R., and G. Imreh. 1997. “Pulling teeth and torture”: Musical theory and problem solv-
ing. Thinking and Reasoning 3 (4):315–36.
———. 2002. Practicing perfection: Piano performance as expert memory. Psychol Sci 13
(4):342–49.
Chaffin, R., and A. F. Lemieux. 2004. General perspectives on achieving musical excellence.
In Musical excellence: Strategies and techniques to enhance performance, ed. A. Williamon.
Oxford: Oxford University Press.
272 Chapter Eight

Charness, M. E., M. H. Ross, and J. M. Shefner. 1996. Ulnar neuropathy and dystonic flexion of
the fourth and fifth digits: Clinical correlation in musicians. Muscle Nerve 19 (4):431–37.
Chen, J., M. Woollacott, and S. Pologe. 2006. Accuracy and underlying mechanisms of shifting
movements in cellists. Exp Brain Res 174 (3):467–76.
Connolly, C., and A. Williamon. 2004. Mental skills training. In Musical excellence: Strate-
gies and techniques to enhance performance, ed. A. Williamon. Oxford: Oxford University
Press.
Delmaire, C., A. Krainik, S. Tezenas du Montcel, E. Gerardin, S. Meunier, J. F. Mangin,
S. Sangla, L. Garnero, M. Vidailhet, and S. Lehericy. 2005. Disorganized somatotopy in the
putamen of patients with focal hand dystonia. Neurology 64 (8):1391–96.
Elbert, T., V. Candia, E. Altenmuller, H. Rau, A. Sterr, B. Rockstroh, C. Pantev, and E. Taub.
1998. Alteration of digital representations in somatosensory cortex in focal hand dystonia.
Neuroreport 9 (16):3571–75.
Elbert, T., C. Pantev, C. Wienbruch, B. Rockstroh, and E. Taub. 1995. Increased cortical repre-
sentation of the fingers of the left hand in string players. Science 270 (5234):305–7.
Ericsson, K. A., R. T. Krampe, and C. Tesch-Romer. 1993. The role of deliberate practice in
the acquisition of expert performance. Psychological Review 100 (3):363–406.
Finney, S. A., and C. Palmer. 2003. Auditory feedback and memory for music performance:
Sound evidence for an encoding effect. Mem Cognit 31 (1):51–64.
Frucht, S. J., S. Fahn, P. E. Greene, C. O’Brien, M. Gelb, D. D. Truong, J. Welsh, S. Factor, and
B. Ford. 2001. The natural history of embouchure dystonia. Mov Disord 16 (5):899–906.
Fry, H. J. 1986. Overuse syndrome in musicians—100 years ago. An historical review. Med J
Aust 145 (11–12):620–25.
Gabrielsson, A. 1999. The performance of music. In The psychology of music, ed. D. Deutsch.
San Diego: Academic Press.
Gaser, C., and G. Schlaug. 2003. Brain structures differ between musicians and non-musicians.
J Neurosci 23 (27):9240–45.
Graziano, M. S., C. S. Taylor, T. Moore, and D. F. Cooke. 2002. The cortical control of move-
ment revisited. Neuron 36 (3):349–62.
Halpern, A. R., and G. H. Bower. 1982. Musical expertise and melodic structure in memory for
musical notation. Am J Psychol 95:31–50.
Hashimoto, I., A. Suzuki, T. Kimura, Y. Iguchi, M. Tanosaki, R. Takino, Y. Haruta, and
M. Taira. 2004. Is there training-dependent reorganization of digit representations in area 3b
of string players? Clin Neurophysiol 115 (2):435–47.
Haslinger, B., P. Erhard, E. Altenmuller, A. Hennenlotter, M. Schwaiger, H. Grafin Von Ein-
siedel, E. Rummeny, B. Conrad, and A. O. Ceballos-Baumann. 2004. Reduced recruitment
of motor association areas during bimanual coordination in concert pianists. Hum Brain
Mapp 22 (3):206–15.
Hassler, M. 2000. Music medicine. A neurobiological approach. Neuroendocrinol Lett 21
(2):101–6.
Haueisen, J., and T. R. Knosche. 2001. Involuntary motor activity in pianists evoked by music
perception. J Cogn Neurosci 13 (6):786–92.
Hirata, Y., M. Schulz, E. Altenmuller, T. Elbert, and C. Pantev. 2004. Sensory mapping of lip
representation in brass musicians with embouchure dystonia. Neuroreport 15 (5):815–18.
Hochberg, F. H., S. U. Harris, and T. R. Blattert. 1990. Occupational hand cramps: Professional
disorders of motor control. Hand Clin 6 (3):417–28.
Hochberg, F. H., and N. S. Hochberg. 2000. Occupational cramps/focal dystonias. In Medical
problems of the instrumental musician, ed. R. Tubiana and P. C. Amadio. London: Martin
Dunitz.
 How the Performance of Music Affects the Brain 273

Hund-Georgiadis, M., and D. Y. von Cramon. 1999. Motor-learning-related changes in piano

players and non-musicians revealed by functional magnetic-resonance signals. Exp Brain
Res 125 (4):417–25.
Hutchinson, S., L. H. Lee, N. Gaab, and G. Schlaug. 2003. Cerebellar volume of musicians.
Cereb Cortex 13 (9):943–49.
Ibanez, V., N. Sadato, B. Karp, M. P. Deiber, and M. Hallett. 1999. Deficient activation of the
motor cortical network in patients with writer’s cramp. Neurology 53 (1):96–105.
Jabusch, H. C., and D. Altenmuller. 2006. Focal dystonia in musicians: From phenomenology
to therapy. Adv Cog Psychol 2 (2–3):207–20.
Jancke, L., G. Schlaug, and H. Steinmetz. 1997. Hand skill asymmetry in professional musi-
cians. Brain Cogn 34 (3):424–32.
Jankovic, J., and H. Shale. 1989. Dystonia in musicians. Semin Neurol 9 (2):131–35.
Kandel, E. R., J. H. Schwartz, and T. M. Jessell. 2000. Principles of Neural Science. 4th ed.
New York: McGraw-Hill, Health Professions Division.
Karni, A., G. Meyer, P. Jezzard, M. M. Adams, R. Turner, and L. G. Ungerleider. 1995. Func-
tional MRI evidence for adult motor cortex plasticity during motor skill learning. Nature
377 (6545):155–58.
Karni, A., G. Meyer, C. Rey-Hipolito, P. Jezzard, M. M. Adams, R. Turner, and L. G. Unger-
leider. 1998. The acquisition of skilled motor performance: Fast and slow experience-driven
changes in primary motor cortex. Proc Natl Acad Sci USA 95 (3):861–68.
Kleim, J. A., T. M. Hogg, P. M. VandenBerg, N. R. Cooper, R. Bruneau, and M. Remple. 2004.
Cortical synaptogenesis and motor map reorganization occur during late, but not early, phase
of motor skill learning. J Neurosci 24 (3):628–33.
Langheim, F. J., J. H. Callicott, V. S. Mattay, J. H. Duyn, and D. R. Weinberger. 2002. Cortical
systems associated with covert music rehearsal. Neuroimage 16 (4):901–8.
Lee, D. J., Y. Chen, and G. Schlaug. 2003. Corpus callosum: Musician and gender effects.
Neuroreport 14 (2):205–9.
Lim, V. K., E. Altenmuller, and J. L. Bradshaw. 2001. Focal dystonia: Current theories. Hum
Mov Sci 20 (6):875–914.
Lim, V. K., J. L. Bradshaw, M. E. Nicholls, and E. Altenmuller. 2004. Abnormal sensorimotor
processing in pianists with focal dystonia. Adv Neurol 94:267–73.
Lotze, M., G. Scheler, H. R. Tan, C. Braun, and N. Birbaumer. 2003. The musician’s brain:
Functional imaging of amateurs and professionals during performance and imagery. Neuro-
image 20 (3):1817–29.
Luders, E., C. Gaser, L. Jancke, and G. Schlaug. 2004. A voxel-based approach to gray matter
asymmetries. Neuroimage 22:656–664.
Maess, B., S. Koelsch, T. C. Gunter, and A. D. Friederici. 2001. Musical syntax is processed
in Broca’s area: An MEG study. Nat Neurosci 4 (5):540–45.
Maguire, E. A., D. G. Gadian, I. S. Johnsrude, C. D. Good, J. Ashburner, R. S. Frackowiak,
and C. D. Frith. 2000. Navigation-related structural change in the hippocampi of taxi drivers.
Proc Natl Acad Sci USA 97 (8):4398–4403.
Meister, I. G., T. Krings, H. Foltys, B. Boroojerdi, M. Muller, R. Topper, and A. Thron. 2004.
Playing piano in the mind—an fMRI study on music imagery and performance in pianists.
Cogn Brain Res 19 (3):219–28.
Morrison, S. 2000. Effect of melodic context, tuning behaviors, and experience on intonation
accuracy of wind players. J Res Music Educ 48:39–51.
Noppeney, U., T. D. Waberski, R. Gobbele, and H. Buchner. 1999. Spatial attention modu-
lates the cortical somatosensory representation of the digits in humans. Neuroreport 10
(15):3137–41.
274 Chapter Eight

Nordstrom, M. A., and S. L. Butler. 2002. Reduced intracortical inhibition and facilitation of
corticospinal neurons in musicians. Exp Brain Res 144 (3):336–42.
Ozturk, A. H., B. Tascioglu, M. Aktekin, Z. Kurtoglu, and I. Erden. 2002. Morphometric com-
parison of the human corpus callosum in professional musicians and non-musicians by using
in vivo magnetic resonance imaging. J Neuroradiol 29 (1):29–34.
Parsons, L. M. 2001. Exploring the functional neuroanatomy of music performance, percep-
tion, and comprehension. Ann NY Acad Sci 930:211–31.
———. 2003. Rethinking the lesser brain. Sci Am 289 (2):40–47.
Pascual-Leone, A. 2001. The brain that plays music and is changed by it. Ann NY Acad Sci
930:315–29.
Pesenti, A., S. Barbieri, and A. Priori. 2004. Limb immobilization for occupational dystonia: A
possible alternative treatment for selected patients. Adv Neurol 94:247–54.
Priori, A., A. Pesenti, A. Cappellari, G. Scarlato, and S. Barbieri. 2001. Limb immobilization
for the treatment of focal occupational dystonia. Neurology 57 (3):405–9.
Pujol, J., J. Roset-Llobet, D. Rosines-Cubells, J. Deus, B. Narberhaus, J. Valls-Sole, A. Cap-
devila, and A. Pascual-Leone. 2000. Brain cortical activation during guitar-induced hand
dystonia studied by functional MRI. Neuroimage 12 (3):257–67.
Pullman, S. L., and A. H. Hristova. 2005. Musician’s dystonia. Neurology 64 (2):186–87.
Ridding, M. C., B. Brouwer, and M. A. Nordstrom. 2000. Reduced interhemispheric inhibition
in musicians. Exp Brain Res 133 (2):249–53.
Romanelli, P., V. Esposito, D. W. Schaal, and G. Heit. 2005. Somatotopy in the basal ganglia:
Experimental and clinical evidence for segregated sensorimotor channels. Brain Res Brain
Res Rev 48 (1):112–28.
Rosenkranz, K., A. Williamon, K. Butler, C. Cordivari, A. J. Lees, and J. C. Rothwell. 2005.
Pathophysiological differences between musician’s dystonia and writer’s cramp. Brain 128
(Pt 4):918–31.
Schlaug, G. 2001. The brain of musicians. A model for functional and structural adaptation.
Ann NY Acad Sci 930:281–99.
———. 2006. Brain structures of musicians: Executive functions and morphological implica-
tions. In Music, motor control and the brain, ed. E. Altenmuller, M. Wiesendanger, and
J. Kesselring. Oxford: Oxford University Press.
Schlaug, G., L. Jancke, Y. Huang, and H. Steinmetz. 1995. In vivo evidence of structural brain
asymmetry in musicians. Science 267 (5198):699–701.
Schmidt, A., H. C. Jabusch, E. Altenmuller, J. Hagenah, N. Bruggemann, K. Hedrich, R. Saun-
ders-Pullman, S. B. Bressman, P. L. Kramer, and C. Klein. 2006. Dominantly transmitted
focal dystonia in families of patients with musician’s cramp. Neurology 67 (4):691–93.
Schneider P., M. Scherg, H. G. Dosch, H. J. Specht, A. Gutschalk, and A. Rupp. 2002. Mor-
phology of Heschl’s gyrus reflects enhanced activation in the auditory cortex of musicians.
Nat Neurosci 5:688–694.
Sergent, J., E. Zuck, S. Terriah, and B. MacDonald. 1992. Distributed neural network underly-
ing musical sight-reading and keyboard performance. Science 257 (5066):106–9.
Sloboda, J. A., J. W. Davidson, D. G. Howe, and D. G. Moore. 1996. The role of practice in
the development of performing musicians. Brit J Psychol 87 (2):287–309.
Sluming, V., T. Barrick, M. Howard, E. Cezayirli, A. Mayes, and N. Roberts. 2002. Voxel-
based morphometry reveals increased gray matter density in Broca’s area in male symphony
orchestra musicians. Neuroimage 17 (3):1613–22.
Sterr, A., M. M. Muller, T. Elbert, B. Rockstroh, C. Pantev, and E. Taub. 1998. Changed per-
ceptions in Braille readers. Nature 391 (6663):134–35.
 How the Performance of Music Affects the Brain 275

Stickgold, R., and M. P. Walker. 2005. Memory consolidation and reconsolidation: What is the
role of sleep? Trends Neurosci 28 (8):408–15.
Taub, E., G. Uswatte, and T. Elbert. 2002. New treatments in neurorehabilitation founded on
basic research. Nat Rev Neurosci 3 (3):228–36.
Tubiana, R. 2000. Musician’s focal dystonia. In Medical problems of the instrumental musi-
cian, ed. R. Tubiana and P. C. Amadio. London: Martin Dunitz.
Turner-Stokes, L., and K. Reid. 1999. Three-dimensional motion analysis of upper limb
movement in the bowing arm of string-playing musicians. Clin Biomech (Bristol, Avon) 14
(6):426–33.
Walker, M. P., T. Brakefield, J. Seidman, A. Morgan, J. A. Hobson, and R. Stickgold. 2003.
Sleep and the time course of motor skill learning. Learn Mem 10 (4):275–84.
Wiesendanger, M., A. P. Baader, and O. Kazennikov. 2006. Fingering and bowing in violin-
ists: A motor control approach. In Music, motor control and the brain, ed. E. Altenmuller,
M. Wiesendanger, and J. Kesselring. Oxford: Oxford University Press.
Williamon, A., and T. Egner. 2004. Memory structures for encoding and retrieving a piece of
music: An ERP investigation. Brain Res Cogn Brain Res 22 (1):36–44.
Williamon, A., and E. Valentine. 2000. Quantity and quality of musical practice as predictors
of performance quality. Br J Psychol 91 (Pt 3):353–76.
———. 2002. The role of retrieval structures in memorizing music. Cognit Psychol 44 (1):1–32.
Wilson, F. R., C. Wagner, and V. Homberg. 1993. Biomechanical abnormalities in musicians
with occupational cramp/focal dystonia. J Hand Ther 6 (4):298–307.
Wynn Parry, C. B., and R. Tubiana. 1998. Dystonia. In The musician’s hand: A clinical guide,
ed. I. Winspur and C. B. Wynn Parry. London: Martin Dunitz.
Yaguez, L., D. Nagel, H. Hoffman, A. G. Canavan, E. Wist, and V. Homberg. 1998. A mental
route to motor learning: Improving trajectorial kinematics through imagery training. Behav
Brain Res 90 (1):95–106.
Zatorre, R. J. 1998. Functional specialization of human auditory cortex for musical processing.
Brain 121 (10):1817–1818.
Zeuner, K. E., W. Bara-Jimenez, P. S. Noguchi, S. R. Goldstein, J. M. Dambrosia, and
M. Hallett. 2002. Sensory training for patients with focal hand dystonia. Ann Neurol 51
(5):593–98.
 Chapter Nine

Hearing and the Processing of

Musical Sound by the Brain

In this chapter we will first consider the physical properties of sound, which are im-
portant for an understanding of how the ear responds to and can be damaged by it. We
will then consider how sound is detected by the ear and how the information it extracts
reaches the brain. Finally, we will examine how music is analyzed by the brain and
how some of the effects that it has on us are generated.

THE NATURE OF SOUND

The sensation that we subjectively perceive as sound originates from objects in the
environment that vibrate. Most commonly, the effects of these vibrations reach us
through the pressure waves that the vibrating objects generate in air, but the auditory
system can be stimulated more directly when the vibrating object (e.g., a tuning fork)
is applied directly to the bones of the skull, and indeed, some hearing aids work on this
principle. In the case of musical instruments, the sound arises from a vibrating string,
the vibrations in a column of air within the instrument, or the membrane of a drum and
is then conducted through the air to the ears. The properties of the vibrations in the
resonating object and those it drives in air are best described initially for a pure tone
(a sound of a single frequency). The sounds of natural objects and musical instruments
are not like this, but pure tones can be generated electronically using a waveform
generator and a loudspeaker. The vibration generating the tone can be represented
graphically (fig. 9-1, sound presentation 9-1) as a regular oscillation that is described
mathematically as a sine wave representing the displacement of the vibrating object
over time. The wave is characterized by its amplitude (which equates to the size of the
displacement of the vibrating cone of the speaker, or the loudness of the sound) and its
frequency (which represents the number of oscillations per second made by the cone
of the speaker), which we may call the pitch of the tone. The vibrations of the speaker
cone set up a series of pressure pulses in the surrounding air. As the cone moves
forward, it produces an increase in pressure as the molecules of the air are pushed
closer together than they would be in the absence of the vibration. As the cone moves

276
 Hearing and the Processing of Musical Sound by the Brain 277

Figure 9-1. When sound is generated by a vibrating body such as the diaphragm of a loud-
speaker, the air is compressed, raising its pressure as the diaphragm moves forward, while
the pressure falls as it moves back. This can be represented by the waveform at the bottom
of the image in which the vertical axis represents air pressure and the horizontal axis, time.
The distance between the peaks is the wavelength. Shortening the wavelength is equivalent
to increasing the frequency, perceived as a rise in pitch. The greater the amplitude of the
waveform, the louder the sound.

backward, the air pressure is reduced, leaving the air molecules farther apart than they
would otherwise be (fig. 9-1). These pressure oscillations radiate out from the origin
of the sound like the ripples in a still pond that spread from the point at which a stone
has hit the water surface and continue for as long as the vibrations persist.
The pressure fluctuations that we perceive as sound can be extremely small. The
pressure of air at sea level, which is generated by the weight of the column of gas that
extends from there to the edge of the atmosphere, provides a standard measurement
that is given the value of 1 atmosphere. It is the same as the pressure that is produced
by the weight of 10 cubic meters of water. The quietest sound that can be detected by
the auditory system is described as its threshold and is represented by a pressure pulse
278 Chapter Nine

that is 2x10-10 (two ten-billionths) of atmospheric pressure. This arbitrary value is used
as the standard sound pressure level (SPL) against which sound intensity or volume
is measured. One of the remarkable features of the auditory system is the enormous
range of sound pressures that can be detected. The pressure generated by sounds at
the top end of the tolerable range (which are perceived as painful) is 107 (ten million)
times that of the auditory threshold. With this range of pressures, it is not convenient
to use a linear scale to describe sound intensities, so a logarithmic scale is used in-
stead. The unit of measure is the decibel (dB), which represents a ratio—in the case
of sound pressure this is defined as follows:

sound pressure level (dB SPL) = 20 log10 (P/Pref),

where P is the pressure of the sound being measured and Pref is the reference sound
level that is the human auditory threshold level described above. The standard units in
which pressure is measured are called pascals (Pa). The standard value assigned to the
auditory threshold is 20 x 10-6Pa (twenty-millionths of a pascal). However, the main
point to bear in mind is that a rise of 20 dB SPL represents a tenfold increase in the
sound pressure. For an orchestra, the typical sound levels during performance range
from about 40 dB SPL for a pianissimo passage to 90 dB for a fortissimo involving
the full orchestra. Players may experience sound pressure levels above this for short
periods. Sound levels of over 100 dB are quite commonly reached for electronically
amplified music at rock concerts and can rise further to levels that cause temporary or
permanent damage to the ear (see below). In the context of its impact on our hearing,
the important characteristic of sound is its intensity, which is the amount of energy
that is released per unit area. This is expressed in units of watts per square meter
(Wm2). As this is proportional to P2:

sound intensity (dB) = 10 log10 (I/Iref),

so a rise of 10 dB represents a tenfold increase, and a 3 dB increase is equivalent to

a doubling of intensity. Thus, between the threshold for hearing (0 dB SPL) and the
intensity of sound that causes pain (about 140 dB SPL) the intensity of the sound (the
amount of energy it carries) increases by a factor of 1014 (one hundred trillion). This
is a quite extraordinary range of intensities; no other sensory system has to cope with
a dynamic range of this magnitude. We will see later some of the mechanisms used
by the ear to achieve this feat.
If the sound is represented by a simple sine wave (fig. 9-1), the frequency of the
oscillation in sound pressure represents its pitch. The standard unit for the measure
of frequency is the hertz (Hz). A frequency of 1 Hz represents a single cycle of the
waveform per second. In musical terms, pitch for Western music is now largely stan-
dardized, with the A above middle C (A4) providing a reference at 440 Hz. Until the
nineteenth century there was no international pitch standard, and in the eighteenth
century A4 typically varied between 400 Hz and 450 Hz, a difference of about three
semitones. The first attempt at pitch standardization in the 1850s set A4 at 435 Hz,
and concert pitch was raised to its present level in the mid-twentieth century. The ten-
 Hearing and the Processing of Musical Sound by the Brain 279

Figure 9-2. This graph shows the intensity threshold of the human ear for sound across the full
range of frequencies that it can detect (20–20,000 Hz). The minimum threshold (defined as 0
dB) is found for sounds of about 3,000–4,000 Hz. Above and below this frequency, the threshold
rises. For the lowest frequencies detectable by the ear, it reaches 80 dB, which is equivalent to the
loudest sound intensity encountered in conversational speech. For reference, the piano keyboard
is drawn at the bottom with the frequency of C indicated for each octave. Its range runs from
about 27 Hz to 4,000 Hz.

dency for pitch to rise was linked to the need to produce a brilliant sound that would
fill large concert halls and was limited mainly by the tensile strength of the E string
of the violin. When it became possible to produce strings that would support greater
tensions reliably, the pitch could be raised, though this also required structural modi-
fications to the violin so that it could support the increased forces this generated. The
change in pitch was perhaps not so welcome to singers, whose instruments cannot be
so easily modified!
280 Chapter Nine

In humans, the audible frequency range extends from about 20 Hz to 20,000 Hz (20
kilohertz [kHz]), though our ability to discriminate pitch (i.e., to distinguish one note
from another) is effectively limited to sounds below 5 kHz. It is important to recognize
that the ear is not equally sensitive to all frequencies (fig. 9-2). Its peak sensitivity lies
in the range of 3–5 kHz, and the threshold for frequencies between about 300–8,000
Hz lies below 20 dB SPL. This covers the range used in conversational speech. Most
significant musical sounds lie below 11 kHz, but fundamental frequencies of the notes
on a full concert piano range from 27 Hz (A0) to 4,186 Hz (C8), with middle C (C4)
at 261 Hz.
Most of the sounds we hear are not pure sounds of a single frequency but are
complex sounds containing many frequencies (sound presentation 9-1). The lowest
frequency of the complex sound produced by an instrument is usually the most intense
and is called the fundamental. This is overlaid by a series of harmonics. If we consider
the vibration of a plucked string, the fundamental (also known as the first harmonic)
represents a vibration of the whole length of the string in a smooth curve from one
end to the other, with the greatest displacement from the rest position at the point of
the string at its midpoint (fig. 9-3). It is quite easy to see this mode of vibration on,
for example, a harp string. What is not so apparent is that superimposed on this is a
further set of vibrations (harmonics). The second harmonic can be visualized as one
in which the midpoint of the string remains at the rest position and the upper and
lower segments vibrate at twice the frequency of the fundamental. This creates a note
one octave above the fundamental that can be used for effect in many stringed instru-
ments by touching the midpoint of the string during plucking or bowing to suppress
the fundamental. The third harmonic is a mode where the string is divided into three
vibrating segments and is a fifth above the second harmonic. For the fourth harmonic,
the string is divided into four segments giving a pitch a third above the previous har-
monic, and so on. When we hear the complex note that results from the sounding of
the fundamental and its harmonics, we nevertheless attribute to it a single frequency
value that is defined as its pitch. This is usually the frequency of the fundamental, but
it may be influenced by the perceived harmonic structure of the note. The perception
of pitch is a reflection of how the brain interprets sound; we will see later that the pitch
of a note may represent a fundamental frequency that is not actually present! In dif-
ferent instruments or voices, the relative strength of the harmonics in the notes varies,
giving each a distinctive sound quality or timbre. For example, among the woodwinds,
the clarinet with its cylindrical bore acts like a stopped pipe and generates only the
odd harmonics of the fundamental; this is a significant element of its characteristic
tone color (Fletcher and Rossing 1998). On the other hand, the oboe, saxophone, and
bassoon, because of their conical bore, act as open pipes and so have both odd and
even harmonics. For brass instruments, a significant component of their timbre comes
from the presence of a mouthpiece and the shape of the bell (Benade 1973). Interest-
ingly, however, much of the information we use to identify the characteristic sound of
a particular instrument comes from the transient changes in the harmonics and other
extraneous sounds that characterize the onset of the note. If we record a note and cut
off the beginning and the end, it becomes almost impossible to identify the instrument
Figure 9-3. The harmonics of a vibrating string. When a string is plucked, a series
of frequencies called harmonics or partials are generated. The fundamental fre-
quency (first harmonic) is the result of the vibration seen in the uppermost panel.
Superimposed on this are the other modes of vibration shown in the lower panels.
These vibrations divide the string into segments of equal length. For the second
harmonic the string is divided into two, and the frequency is twice that of the funda-
mental; for the third harmonic the frequency is three times that of the fundamental,
and so on. This type of harmonic series is typical of all real sounds.
282 Chapter Nine

that made it. One cannot even say whether the sound comes from a stringed or a wind
instrument. This is illustrated in sound presentation 9-1 on the disc accompanying the
text. When the onset of the note is restored, the task becomes easy.
The auditory system can be divided into two parts, which we will consider in turn.
The first is the ear, which detects the sound and splits it up into its frequency compo-
nents. This information is then carried to the second part, which lies within the brain;
it is here that the frequency components are recombined, analyzed, and interpreted.
The brain also compares the information coming from the two ears to create a three-
dimensional soundscape, which allows the position of the sound source to be accu-
rately determined, at least for some frequencies.

HEARING AND THE EAR

The Outer and Middle Ear

For descriptive purposes, the ear is usually divided into three parts, the outer,
middle, and inner ear (fig. 9-4A). The outer ear is composed of the cartilaginous
external ear, which sits on the side of the head, and the auditory canal into which
it reflects sound. The external ear is most effective at collecting sound from the
front and so has a built-in directionality. Unlike the ears of some other animals such
as the cat, our external ears are virtually immobile, though the muscles that move
them still exist as rudiments. The auditory canal ends in the eardrum (or tympanic
membrane), which separates it from the air-filled middle ear. Like the membrane
of a drum, the tympanic membrane must be kept taut if it is to vibrate efficiently in
response to sound. On the other side of the tympanic membrane lies the middle ear
(fig. 9-4B). The vibrations of the tympanic membrane are communicated to three
small bones (or ossicles) that lie in contact with each other within the middle ear
cavity. These are called the malleus (hammer), incus (anvil), and stapes (stirrup).
The malleus, as its name suggests, is shaped like a hammer with the handle lying in
contact with the tympanic membrane. A muscle called the tensor tympanum acts on
the handle to keep the tympanic membrane taut. This muscle is supplied by a branch
of the trigeminal nerve, which also supplies the large chewing muscles that move
the jaw. Vibrations of the tympanic membrane are conducted through the malleus,
incus, and stapes to the fluid-filled inner ear, where the structures that detect the
vibrations are found. The middle and inner ear lie within one of the bones of the
skull (the temporal bone). There are two windows in the bony wall separating the
inner ear from the middle ear, one of which is round and the other oval (fig. 9-4B).
The stapes ends in a flat disk called the footplate, which sits in the oval window.
Because the fluid in the inner ear (which is mainly water) is incompressible, vibra-
tions of the footplate would be unable to set up vibrations in it if it were a sealed
chamber. The second window (the round window) is covered with a membrane,
and as the footplate pushes into the oval window, this membrane bulges outward.
This allows the vibrations to radiate through the fluid. The ear ossicles are not there
simply to carry the vibrations across the middle ear chamber; they also concentrate
the energy of the sound that is vibrating the eardrum onto the much smaller area of
 Hearing and the Processing of Musical Sound by the Brain 283

Figure 9-4. The structure of the ear. A. The ear can be divided into three distinct regions. The
outer ear consists of the cartilaginous part on the outside of the head and the auditory canal into
which it reflects sound. The canal ends in the tympanic membrane, behind which lies the middle
ear. This is an air-filled cavity containing three small bones (ossicles) called the malleus (ham-
mer), incus (anvil), and stapes (stirrup). The middle ear chamber is connected to the back of the
throat by the Eustachian tube to allow pressure equalization. The inner ear is made up of a series
of convoluted spaces filled with fluid. These are the cochleas, which respond to sound, and the
semicircular canals and vestibule, which are involved in balance. B. The middle ear. The tympanic
membrane is attached to the “handle” of the malleus and is kept taut by the action of the tensor
tympanum muscle. The stapes ends in a broad “footplate” that is pushed into the oval window,
a hole in the bone that contains the inner ear. The movement of the stapes can be limited by the
stapedius muscle, whose tendon attaches to its head.

the oval window. The ossicles also act as a system of levers, further contributing
to this effect so that there is an increase in the force exerted per unit area of about
twenty times. This amplification is necessary because the airborne vibrations of
the tympanic membrane would otherwise be too weak to drive the vibrations in the
fluid of the inner ear; in fact without this mechanism, most (over 99 percent) of the
energy would be reflected from the surface of the fluid.
Figure 9-5. A–B. Protecting the ear. The stapedius muscle limits the amplitude of the vibrations
of the stapes and so the amount of energy this transfers to the inner ear. It is more effective in
blocking low-frequency than high-frequency sound. When a high-pitched and a low-pitched
sound are presented together, each sets up a traveling wave that runs along the basilar membrane
(solid line). Without the stapedius muscle, the wave produced by the low-frequency sound (dark)
can mask that of the high-frequency sound (light), making it impossible to detect. When the
stapedius is active, the greater reduction in the low-frequency wave allows the high-frequency
sound to emerge. The resultant waveform perceived by the cochlea is shown on the right. C–D.
Because the physical properties of the basilar membrane change along its length, waves caused by
high-frequency sounds travel only a short distance before being damped, while successively lower
frequencies generate waves that travel ever farther. How far the wave runs therefore indicates
its frequency. This is the principle of tonotopy, and this is one source of information that can be
used by the nervous system to determine the pitch of the sound.
 Hearing and the Processing of Musical Sound by the Brain 285

The movement of the stapes is controlled by the stapedius muscle, which is attached
to its neck (fig. 9-4B). When we are exposed to very loud sounds, the muscle contracts
and so limits the extent to which the stapes can vibrate, which helps to protect the deli-
cate receptor apparatus of the inner ear. The contraction of the muscle is brought about
by an involuntary reflex (the stapedius reflex) (fig. 9-5A, B). The stapedius restricts
the movement of the chain of ossicles, with the result that they transmit high frequen-
cies more efficiently than low frequencies. When the stapedius is active, the chain of
ossicles therefore acts as a filter, making it easier to detect a high-frequency sound
against a low-frequency background. The muscle becomes active just before we speak
or sing. During speech the sound of one’s own voice is transmitted to the ears not
only through the air, but also through the tissues of the neck and head, which most ef-
fectively transmit low-frequency vibrations. This is one reason why our voice sounds
very different when we listen to it in a recording. The volume of the voice is also
normally set at a level that will carry the sound at a comfortable volume to a listener
who typically is much farther away from us than the distance between our mouth and
our own ears, and so it is appropriate to activate the reflex to reduce the sensitivity of
our hearing during speech. The stapedius muscle is supplied by the facial nerve, which
also supplies the small muscles in the face that are used to generate facial expression
or to control the embouchure. This nerve is sometimes damaged in a condition known
as facial nerve palsy (Bell’s palsy), which causes paralysis of the facial muscles (see
chapter 6). One symptom of facial nerve palsy is an intolerance of loud noise, which
is a direct consequence of the paralysis of the stapedius muscle.
Like all gases, air changes its volume depending on its pressure. The air in the
middle ear is no exception. We experience changes in atmospheric pressure constantly
as weather systems pass over or if we climb a hill or a mountain. During flying or
scuba diving, the pressure changes can be both rapid and dramatic. The cabin of a
commercial passenger aircraft is pressurized because at the cruising height for a typi-
cal long-haul flight (a mile or more higher than Mount Everest), we would rapidly
become unconscious due to low oxygen pressure. However, the cabin pressure is
usually set at a level equivalent to about 6,000 feet above sea level, which is why we
still have to make adjustments during takeoff. If the pressure in the middle ear is not
the same as the outside air, the eardrum will bulge inward or outward. At best this is
uncomfortable, but as the pressure difference increases, it rapidly becomes painful and
the eardrum will ultimately rupture. Though this would normally heal spontaneously
over a couple of months, hearing would be affected until the process is complete.
There must therefore be a link between the middle ear and the outside air to allow
pressure equalization; this duct, which links the middle ear to the back of the throat,
is the Eustachian or auditory tube (fig. 9-4B). For part of its length it travels through
bone, but the end of the tube is made of cartilage. Most of the time this section is
closed, but part of the muscle layer lining the inside of the throat is attached to it and
during swallowing, the contraction of the muscle opens the tube briefly. This is why
swallowing or any activity that promotes it (e.g., sucking on a sweet) is encouraged
by cabin staff during takeoff. The eardrum may also rupture following an explosion,
where the increase in air pressure occurs too rapidly for pressure equalization to take
place via the eustachian tube.
286 Chapter Nine

The Inner Ear

The inner ear is composed of a convoluted chamber lying within the temporal bone
of the skull. Because of the mazelike complexity of its passageways, it is known as
the labyrinth. One part of the inner ear contains the sensory apparatus that monitors
the orientation and acceleration of the head, information that the brain uses to help
us maintain our balance. This system does not concern us here. The other part, where
sound is detected, is called the cochlea (fig. 9-6). This is a spiral of two and three-
quarter turns with a central bony core, and a cast of it would look rather like a snail
shell; indeed, the name cochlea is derived from the Latin word for “shell.” The base of
the cochlea opens into a large space called the vestibule, which has the oval window in
its wall (fig. 9-4B). A cross-section through the cochlea reveals that it contains three
long, narrow, fluid-filled chambers running in parallel up to the tip of the spiral. Each
of these spaces is given the name “scala.” This comes from the Latin word for “ladder”
or “staircase” and has the same origin as the musical scale. The central tube, called
the scala media (or cochlear duct), comes to a blind end at the apex of the cochlea.
On either side are the scala vestibuli and the scala tympani, which communicate with
each other at the tip of the spiral. The floor of the scala media is formed by the basilar
membrane (figs. 9-5C, D, 9-6). This is a complex structure whose physical properties
are responsible for separating complex sounds into their different frequency compo-
nents. It is about 3.5 millimeters long and is relatively narrow and stiff at the base of
the spiral but broad and floppy at the apex. The vibrations in the fluid cause traveling
waves to run along the membrane. To understand what is meant by a traveling wave,
lay a piece of rope on the ground and whip one end up and down; a wave will run
along the rope to its other end. This is analogous to what happens with the basilar
membrane, but how far the wave travels depends on its frequency. High-frequency
sounds run well where the membrane is stiff but die out as it becomes floppier toward
the tip of the cochlea. The lower the frequency of a sound, the farther it can run along
the membrane. As it travels, the wave increases to a peak then abruptly collapses as
the propagation fails, so with increasing frequency the peak occurs farther and farther
along the basilar membrane. The frequency of a harmonic can therefore be identified
by the position of the peak on the membrane. This principle is called tonotopy (topos
is the Greek word for “place”) and has an important influence on the architecture of
the auditory pathways linking the cochlea to the brain.
The vibrations of the basilar membrane are detected by the organ of Corti, a long,
narrow structure composed of sensory cells (hair cells) and nerve endings that runs
along its surface within the scala media (fig. 9-6B). It is named after the micros-
copist Marchese Corti, who first described its structure in 1851. The flask-shaped
hair cells are held upright by specialized support cells (fig. 9-7A). Hair cells are so
named because on their upper surface they have three rows of hairlike projections
(stereocilia) that project into the fluid of the scala media (figs. 9-6, 9-9A). These lie
in contact with a gelatinous flap called the tectorial membrane that cantilevers out
from the bony core of the cochlear spiral. When the basilar membrane vibrates, the
hair cells move relative to the tectorial membrane and the stereocilia are deflected
from side to side.
 Hearing and the Processing of Musical Sound by the Brain 287

Figure 9-6. A. The cochlea is a spiral made up of three parallel tubes running around a core of
bone that contains the auditory nerve. The three tubes are the scala vestibuli and scala tympani
and the cochlear duct (scala media) lying in between. The cells that detect the vibrations caused
by sound are found in the organ of Corti, which sits on the basilar membrane forming the floor
of the cochlear duct. B. The organ of Corti contains four rows of hair cells, which detect vibra-
tions. When the basilar membrane moves up and down, the “hairs” on the tips of these cells are
pushed from side to side by a gelatinous flap called the tectorial (roof) membrane. The hair cells
send signals to nerve cells, which carry information to the brain.

The three rows of stereocilia on each hair cell are not the same height. The row
nearest the edge of the cell is tallest, the middle row is shorter, and the inner row,
shorter still. The tips of adjacent stereocilia in the three rows are connected by fine
strands called tip links. When the stereocilia bend in the direction of the tallest row,
the distance between the tips increases and the tip links are stretched. When they bend
in the opposite direction, the tip links relax. The increase in tension is transmitted
to the membrane of the stereocilia, where it causes minute pores or channels in the
membrane to open. This allows potassium ions to flow into the hair cells from the
fluid in the scala media. As a result, an electrical potential is generated within the hair
cell that causes it to release a neurotransmitter (glutamate) to branches of nerve cells
288 Chapter Nine

that lie in the core of the cochlear spiral. As a result, a series of electrical impulses run
along their axons and into the central nervous system. The hair cells are phenomenally
sensitive to vibration, and at the threshold for hearing, the deflection of the stereocilia
is no greater than the diameter of a single atom.
There are four rows of hair cells, a single inner row that lies closest to the central core
of the cochlea, and three outer rows (figs. 9-6, 9-9A). These are ranged on either side
of a tunnel that passes along the length of the organ of Corti. The inner and outer hair
cells have very different functions. The inner hair cells have abundant connections with
nerve cells, and despite being in the minority, they are mainly responsible for detecting
the vibrations generated by sound. The outer hair cells have much sparser connections
with nerve cells, and their main role is to increase the sensitivity of the organ of Corti.
When the outer hair cells are excited by the movement of the basilar membrane, they
respond by actively contracting (becoming shorter) as the stereocilia move in one direc-
tion, then relaxing (elongating) as they move back in the opposite direction (fig. 9-7B).
The length of the cells oscillates in time with the rise and fall of the basilar membrane,
and this increases the amplitude of the movement in the same way that a gymnast on a
trampoline rises ever higher by jumping in phase with its bounce. The maximum height
of the traveling wave is increased by about one hundred times by this mechanism (fig.
9-7C, D). This activity by outer hair cells constitutes what is known as the cochlear
amplifier. The vibration it induces in the basilar membrane can cause it to operate like
the diaphragm of a loudspeaker so that it can actually generate sound. These are called
otoacoustic emissions and are usually too weak to be audible over ambient noise; how-
ever, in very quiet surroundings, or if the cochlea has been damaged by loud noise or
disease, they may become audible. One famous and well-verified story of otoacoustic
emissions concerns a man who noticed that a dog sitting next to him on a sofa appeared
to be humming. On further investigation he was able to determine that the sound was
coming not from the dog’s throat, but from one of its ears!

Tartini Tones
The cochlea is not only sensitive to sounds that exist in the outside world, but can
also give the illusion of sounds that do not actually exist. This was first described in
the eighteenth century by the violinist and composer Giuseppe Tartini. An analysis
of Tartini’s theories on this and other matters can be found in Planchart (1960). He
found that when two notes are played simultaneously, other notes may also be heard
at mathematically related pitches. These are referred to as Tartini or difference tones.
They are best achieved with relatively low intensity sounds of about 50–65 dB, a
sound level that is comparable to the volume of conversational speech. If the two real
sounds have frequencies of f1 and f2 (where f2 is the higher frequency), the Tartini
tones will have frequencies of f2 – f1, f2 + f1, 2f1 – f2, 2f2 – f1, and so on. An example can
be heard in sound presentation 9-2 on the accompanying disc. This is a phenomenon
well known to organ builders, who have used it for well over a century to generate
what is known as the Acoustic Bass stop. The name is more than a little ironic as the
Acoustic Bass stop gives the impression of a set of pipes that do not exist, but instead
is an illusion created entirely within the auditory system. For example, combining C3
Figure 9-7. A. A hair cell. The movement of the basilar membrane is detected by the hair
cells (see fig. 9-6B). These are supported by phalangeal cells, which have a strutlike exten-
sion that helps to hold them upright. The tips of the cells carry three rows of “hairs” (or
stereocilia). When these are deflected one way, the hair cells are excited, and when they
are moved in the other direction, the cells are inhibited. When excited, their synaptic ves-
icles release a neurotransmitter that stimulates the branches of the nerve cells that carry
the information to the brain. The responsiveness of the hair cells can also be controlled
by signals traveling in the opposite direction, from the brain. B–D. The extreme sensitiv-
ity of the cochlea is partly due to a mechanism called the cochlear amplifier. When the
outer hair cells are activated by the deflections of their stereocilia, they respond by rapidly
shortening and lengthening in phase with the vibrations that excited them (B). This in-
creases the amplitude of the traveling wave. C represents diagrammatically the amplitude
of the wave in the absence of the cochlear amplifier, and D when the cochlear amplifier is
active, though the difference in the amplitude of the wave is much greater in reality.
290 Chapter Nine

at 126 Hz, played on a four-foot-long organ pipe, with G3 at 190 Hz on a pipe that is
two feet eight inches long, can be used to generate the illusion of C2 at 64 Hz (190
– 126 = 64, i.e., f2 – f1) without the need to build an eight-foot C2 pipe to produce it.
In this example, the fundamentals of the notes played represent the second and third
harmonics of the Tartini tone frequency and so when heard together, the three will
give the impression of a single pitch, although this will be less rich than that produced
by an actual eight-foot pipe. In this example, the ratio of f2 to f1 is 1.5, close to the
optimum for producing the effect, which is about 1.2. (Further details of the use of
Tartini tones in the organ can be found in Bicknell 1999 and Audsley 1965). Outside
the arcane world of organ building, the role of Tartini tones in music is a relatively
minor one, as they are generally rather weak, though some violinists use them to con-
trol the intonation of double-stopped notes (Rasch and Plomp 1999).
It is possible to examine the movement of the basilar membrane as a note of a
particular frequency is sounded. If a pure sound of a single frequency is presented,
then one peak of vibration is seen on the membrane. If two pure tones are presented,
we would intuitively expect to see two peaks; however, if the frequencies of the tones
strongly stimulate the same section of the organ of Corti, several additional peaks
may appear, each representing the frequency of a Tartini tone. The Tartini tones are
thought to originate from the outer hair cells as a consequence of changes in the elas-
ticity of the hair bundles when the tip links flip the potassium channels between their
open and closed states (Jaramillo, Markin, and Hudspeth 1993). This disturbs the oth-
erwise linear relationship between the degree of deflection of the stereocilia and the
force they generate—a force that acts to return them to the rest position. The response
is therefore said to be nonlinear, and the resulting wave on the basilar membrane is
called a distortion product. The nonlinear response introduces an instability into the
backward and forward movements of the stereocilia, which either directly, or through
the contractile mechanisms of the outer hair cells, causes the basilar membrane to vi-
brate at the new frequencies. This generates traveling waves that run along the basilar
membrane in exactly the same way that those generated by external sounds do. They
are therefore interpreted by the nervous system as sound even though they have no
reality in the external world. Blocking the potassium channels eliminates the Tartini
tone responses but has no effect on the vibrations generated by the two real tones used
to create them. Nevertheless, the vibrations of the basilar membrane that underlie Tar-
tini tones can sometimes give rise to real sounds in the form of otoacoustic emissions.
While the Tartini tones important in music are usually at a frequency equal to f2 – f1, a
strong otoacoustic emission can, with suitable equipment, be generated at a frequency
of 2f1 – f2. This is known as a distortion product otoacoustic emission (DPOAE).
These can be used clinically to investigate the health of the inner ear (particularly the
functional integrity of the outer hair cells) and are useful in patients such as infants
who are unable to give a clear voluntary response.

Control of Hair Cell Activity by the Central Nervous System

Microscopic examination of the synaptic contacts at the base of the hair cells reveals
that they not only send information to the brain, but also receive it. This allows the
 Hearing and the Processing of Musical Sound by the Brain 291

brain to exert a degree of feedback control over hair cell responsiveness (Brown
1999). The contacts (fig. 9-7A) are found mainly on the outer hair cells that lie within
the sections of the organ of Corti that are responsive to mid- to high-frequency sound.
The nerve cells responsible lie in a region of the brain stem called the superior olivary
complex, and they effectively turn down the volume control on the cochlear amplifier.
This is probably a protective mechanism to prevent damage to the hair cells by high-
intensity sounds, as the nerve cells are most active when such conditions are experi-
enced. When two sounds of different frequencies occur simultaneously, the vibrations
of the basilar membrane created by one can by various means obscure those generated
by the other. The feedback from the brain onto the outer hair cells can also reduce this
interference by toning down the response to the louder frequency.
The superior olivary nucleus is also capable of influencing the signals carrying
pitch information from the inner hair cells. It does this not by direct connections with
the hair cells, but by acting on the neurons within the cochlea that receive signals from
them. Because of the technical difficulties of studying such connections, much less is
known about this system; however, it appears to embody two types of feedback con-
trol. One of these would have a blanket effect across a large proportion of the auditory
range, while the other would be capable of the selective control of narrow frequency
bands (Warr and Boche 2003). Though there is currently no direct evidence for this,
such a system might help us to concentrate on sound frequencies characteristic of
a particular instrument against the background of an orchestra, or on one particular
voice in a room full of people. However, if it does play such a role in selective atten-
tion, this feedback is probably only part of the mechanism involved, as the most com-
plex analysis of sound (e.g., the recognition of the harmonic structure characteristic of
the sound of a particular instrument or voice) is only possible within the cortex.

PROBLEMS AFFECTING THE EAR

Hearing impairment is clearly a great worry for musicians. Most hearing problems origi-
nate from the ear and not from the processing of sound by the nervous system. There are
two major categories of hearing loss—conductive deafness and nerve deafness. Conduc-
tive deafness occurs as a result of damage to or changes in the properties of the eardrum
or ear ossicles. The small moving parts in the middle ear are extremely vulnerable and
may be affected by many diseases or sources of physical damage. Though several re-
markable surgical procedures are now available to treat some of these conditions and
offer an improvement, they generally cannot completely restore the level of hearing to
its original state. By contrast, neural deafness is caused by damage to the hair cells, the
auditory nerve, or the auditory pathways in the brain and is irreversible.

Problems Affecting the Eardrum

Anything that interferes with the vibration of the tympanic membrane will reduce the
keenness of hearing. Getting water in the ear after swimming is an example that most of
us are familiar with. This quickly resolves itself as the water evaporates, but a buildup
292 Chapter Nine

of earwax can cause similar symptoms. The wax is secreted by glands in the external ear
canal. Its purpose is to protect the skin lining the canal from moisture and to discourage
fungal or bacterial growth. The idea that because of its bitter taste (take my word for
it!) this wax acts as a deterrent to insect invasion of the ear canal is still perpetuated in
some medical textbooks, but there appears to be no objective evidence to back it up. The
notion relies on the insect world having similar tastes to ours, and a moment’s reflection
will make it clear that this is often not the case. In a blocked ear, the wax may become
hard, but it can be softened by putting a few drops of light vegetable oil into it for a few
days before washing it out with a large syringe (without a needle!) containing warm
water. The movement of the tympanic membrane and of the ear ossicles can also be
impeded by mucus, which may build up inside the middle ear as a result of an ear infec-
tion (e.g., otitis media). This is most common in children, in whom it is often known as
glue ear, but is sometimes experienced by adults. If the mucus cannot drain effectively
through the Eustachian tube, a grommet (a small rubber tube) may be inserted through
the eardrum for this purpose. The grommet ultimately drops out and the hole heals spon-
taneously without further treatment.

Problems Affecting the Ear Ossicles

The free movement of the ear ossicles can be impeded by abnormal bone growth in a
condition known as otosclerosis. It is usually the stapes that is most affected, losing its
ability to move freely in the oval window. Otosclerosis can have a genetic component
and may therefore run in families, though the hearing may not be noticeably affected in
all individuals who have it. It is twice as common in women than in men, and the symp-
toms (hearing loss sometimes accompanied by dizziness and tinnitus [see below]) may
become more noticeable during pregnancy or at menopause. It is now frequently treated
by surgery to replace the damaged stapes with an artificial one made of Teflon.

Drug-Induced Hearing Loss

Some commonly used prescription drugs may cause tinnitus or damage hair cells in
the cochlea (leading to deafness) or vestibular system (involved in balance) of the
ear. These include some classes of antibiotics (particularly, but not exclusively a
group called aminoglycosides), the antimalarial drug chloroquine, some nonsteroidal
anti-inflammatory drugs (including aspirin in high doses), and certain drugs used for
treating fluid retention. The effects are usually dependent on the dose and length of
time over which the drug is taken. Some of the symptoms of drug toxicity may be
reversible (e.g., tinnitus).

Tinnitus
The word tinnitus literally means a ringing sound, and though this is a common
symptom, the term is used to cover a wide variety of sounds (clicking, high-pitched
whistling, rhythmic or intermittent roaring, etc.) that do not originate from the out-
side world. It is quite often associated with a hypersensitivity to sound (hyperacusis).
 Hearing and the Processing of Musical Sound by the Brain 293

Though it can be a symptom of a wide variety of auditory problems including deaf-

ness, most of us would notice mild tinnituslike symptoms if we were to be placed in
a soundproof room or if our perception of external sounds was reduced because the
movement of the tympanic membrane or ossicles was limited by earwax, glue ear, or
otosclerosis. This is because some forms of tinnitus represent real sounds that occur in
the head. These may result from the flow of blood through vessels in and around the
ear, spasms of the muscles of the middle ear, or movement of the jaw joint. The pres-
ence of this type of sound can sometimes be confirmed by a doctor using a stethoscope.
In other circumstances, tinnitus may be an auditory illusion created by damaged hair
cells (e.g., following exposure to loud noise) or by tumors in the ear or central nervous
system. Because it has so many possible causes, tinnitus requires careful investigation;
however, in many cases an obvious cause cannot be established. Where there is a clear
organic cause this can be addressed directly. Treatment may be as simple as remov-
ing earwax or changing long-term medication. More rarely, delicate surgery on blood
vessels in the ear or treatment of tumors may be required. In most cases, however, the
approach taken is to help the patient cope with the symptoms. Masking the tinnitus by
listening to music or by generating sound in the ear canal may be effective in reduc-
ing its annoyance. A more recent development is tinnitus retraining therapy, which
encourages the brain to become used to the tinnitus to the extent that it can be ignored.
This is called habituation and is an important mechanism by which the nervous system
disregards much of the sensory input it receives so that it does not become overloaded
with information of little significance. For example, in an office near a busy road, we
quickly cease to be aware of the traffic noise when working, just as we cease to notice
the contact of our clothes with the skin, though a new stimulus such as a light touch
on the hand will be recognized immediately.
As tinnitus is a common condition, it is not surprising that many famous musicians
(including Beethoven and Schumann) have suffered from it; however, it is perhaps
most dramatically exemplified by Smetana. In the mid-1870s he became deaf, prob-
ably as a result of syphilis, and complained of a piercing whistling sound in his ears. In
his autobiographical first string quartet, “From My Life,” he illustrates this in the final
Vivace movement by having a high E played continuously for several bars (224–230)
by the first violin (fig. 9-8). Lasting as it does for only a few bars, this must rank as a
remarkably restrained evocation of his symptoms (though in the preceding section, the
E an octave below struggles to establish itself over a more tumultuous background).
Many quartets today play the high E with such a fine sense of musicality that it loses
much of its sinister nature and hardly does justice to the distress it caused Smetana.
The extract from this movement, recorded specially for the accompanying disc by the
Mavron Quartet, presents a more dramatic evocation of the sound that blighted the last
few years of Smetana’s life.

Menière’s Disease
Menière’s disease is caused by a swelling of the part of the ear involved in balance.
This produces the symptoms of dizziness or vertigo, sometimes associated with
unusual movements of the eyes, which use information from this part of the ear to
Figure 9-8. Tinnitus (a ringing, whistling, or growling noise originating within the ear) has
been experienced by many composers. Smetana illustrates this musically in the final move-
ment of his string quartet “From My Life.” In the meno presto, a persistent high E over a
rumbling, monotonous bass briefly (and with commendable restraint) represents what the
phenomenon sounds like. The accompanying disc contains a recording of part of this move-
ment made by the Mavron Quartet.
 Hearing and the Processing of Musical Sound by the Brain 295

compensate for movement. Nausea and sweating may be experienced. Disturbances

in hearing also frequently occur, including loss of sensitivity to low frequencies in the
affected ear and tinnitus. The immediate cause is unknown but it can be triggered by
middle ear, respiratory, or viral infections, head trauma, adverse reactions to prescrip-
tion drugs, alcohol, or stress.

Age-Related Hearing Loss (Presbycusis)

As we age we tend to experience a progressive loss of hearing, starting quite young
with the highest frequencies of the auditory range. If you remember the squeaks of
echolocating of bats in your youth and wonder why they now appear to have fallen
silent, this is the reason. Age-related hearing loss affecting less extreme parts of the
range generally first becomes noticeable at around the age of fifty. By age sixty-five,
approximately 25 percent of people will experience it to some degree and by seventy-
five to eighty, 40–50 percent are affected. The reasons for age-related hearing loss are
unclear, though there appears to be a genetic predisposition for it in some families.
One suggestion is that it is the result of the accumulated effects of a lifetime’s expo-
sure to noise; however, there is no clear evidence for or against this. It is, however,
probably a mistake to consider it to be a single, well-defined condition, as a whole
range of problems (e.g., otosclerosis described above, exposure to loud noise) may
contribute to its onset. There is no cure, though specific problems such as otosclerosis
may be amenable to treatment.

Noise-Induced Hearing Loss

Perhaps of greatest relevance to musicians is the possible risk of noise-induced hearing
loss. Protective mechanisms such as the stapedius reflex are effective in defending the
ear against sounds only up to a certain level. When these defenses are overwhelmed,
physical damage to the hair cells occurs, ultimately destroying the stereocilia and
killing the hair cells (fig. 9-9). While this was once considered to be due directly to
physical forces acting on the stereocilia (e.g., their forceful impact with the tectorial
membrane), it now appears that its major cause is an overloading of the metabolic
pathways of the hair cells. An excellent account of current thinking on the origin of
noise-induced hearing loss can be found in Henderson et al. (2006). A detailed exami-
nation of the sound-abused ear reveals that the damage is not solely confined to the
hair cells, but can involve the other classes of cell within the organ of Corti that help
to maintain a stable environment for them. Exposure to high-intensity sound has the
additional consequence of reducing the blood flow to the cochlea during the period
when the hair cells are at their most active and therefore have their highest demand
for oxygen and nutrients. One consequence of the active state of the hair cells is that
there is a buildup of molecules called reactive oxygen species that oxidize and dam-
age cellular components such as the lipids that form the cell membrane. This initially
has a disruptive effect on the structure of the stereocilia, breaking the tip links that
are necessary for their responsiveness and making them lose contact with the grooves
in the tectorial membrane into which they are normally inserted. It is the outer hair
296 Chapter Nine

Figure 9-9. This figure shows high-magnification images of normal

and damaged hair cells in the organ of Corti obtained using a scan-
ning electron microscope. A. A view looking down onto the tips of
hair cells in the normal cochlea reveals the “hairs” (stereocilia) that
detect movement of the basilar membrane during hearing. Three
rows of outer hair cells and one row of inner hair cells are clearly
visible. B. Prolonged exposure to very loud noise or to certain drugs
damages the “hairs,” especially those of the outer hair cells, as
shown in this photograph. This destroys the cochlear amplifier and
greatly reduces hearing sensitivity. Photographs C–E show progres-
sively more severe damage to the “hairs.” In C they are only slightly
distorted while in D, most have lost their tips and only a small tuft
on the left appears normal. In E, all that remains of the “hairs” is a
fused mass that will be totally unresponsive to sound. (Reproduced
with permission from R. V. Harrison, Auditory Science Laboratory
of the Hospital for Sick Children, Toronto, Canada).

cells that are most affected because their metabolism must support the energetically
demanding cellular contractions that underlie the cochlear amplifier. Any impairment
of their activity results in marked reduction in hearing sensitivity. The effects are at
first reversible, causing what is known as a temporary threshold shift in hearing. After
being at a loud concert, we may feel our hearing is muffled in the following hours or
even days, and this can make speech comprehension difficult. It may also be accom-
panied by other symptoms such as tinnitus. If the damage is not too severe, the hair
cells gradually recover. The stereocilia may return to normal and insert once more
into the pits on the tectorial membrane, and hearing sensitivity is restored. However,
prolonged exposure to intense noise results in a permanent threshold shift (a loss of
hearing sensitivity) as a result of the death of the hair cells. Unfortunately, these never
regenerate in humans, though in some animals such as birds this does occur. The outer
 Hearing and the Processing of Musical Sound by the Brain 297

hair cells at the basal (high frequency) end of the cochlea are the first to be affected,
but ultimately the damage spreads to the other regions of the cochlea and may affect
the inner hair cells as well.
The damage caused by high-intensity sound is frequency specific because the
region of the basilar membrane showing the greatest amplitude of vibration is depen-
dent on frequency. Before the advent of occupational health legislation that made ear
protection mandatory, sound-induced deafness was a major hazard among industrial
workers. It was often known as “boilermaker’s deafness” and was particularly serious
if it affected the frequency range most used in speech (500–4,000 Hz). Noise-induced
hearing loss is frequently first observed between 3,000 and 6,000 Hz. This loss at
the upper end of the main speech range makes consonants difficult to distinguish.
Though industrially induced hearing loss is now rarely seen in the West, the rise of
the personal stereo has caused a new wave of noise-induced deafness. Because music
usually covers most of the audible frequency range, the effects are much less selec-
tive than those of industrial deafness. In addition, the gradual onset of deafness may
go undetected, and the natural reaction is to progressively increase the volume in the
headset to compensate for it, which ultimately exacerbates the problem.
Music-induced deafness among rock musicians, who use electrical amplification
to very high levels (often peaking at over 120 dB), is unsurprising, though they may
be more able to take preventive action by moving about the stage (e.g., to position
themselves out of the main line of fire of the speakers), and be quite content to wear
visible ear protection. A general awareness that classical musicians (particularly those
working in orchestras) may also be exposed to damaging levels of sound has been
slower to gain acceptance, though it is unlikely to come as a surprise to the musicians
themselves! It is not only the absolute intensity of ambient noise that needs to be taken
into consideration, but also how efficiently different frequencies are transmitted by
the ear canal, that is, its resonance characteristics. The intrinsic resonance of the ear
canal lies between 2 and 4 kHz, with a peak at around 3 kHz. The enhancement of
these frequencies due to ear canal resonance ranges from less than 10 dB to nearly 30
dB. This can result in a peak in intensity within the ear at around 3 kHz even though
the intensity envelope of the sound, as measured externally, is actually falling at this
frequency (Henoch and Chesky 1999). Audiograms of patients with noise-induced
hearing loss generally first show a reduction in sensitivity in this region. The degree
to which sound exposure for a particular musical note is increased by ear canal reso-
nance varies from instrument to instrument, as it depends not only on the intensity of
the fundamental frequency, but also on the strength of the harmonics. For example,
the intensity spectrum for both G4 (392 Hz) on the trumpet and G5 (783 Hz) on the
oboe peak at about 3–3.5 kHz, as G7 at 3,136 Hz is a strongly projected harmonic of
these notes in each case.
A useful review of current literature on hearing loss in musicians, providing a criti-
cal assessment of the problems that exist with much of this literature, can be found
in Behar, Wong, and Kunov (2006). Though many studies have sought evidence for
music-related hearing loss in classical musicians (Sataloff 1991; Palin 1994), no clear
picture has so far emerged. There are a number of possible reasons for this. First, the
numbers of musicians examined in each study are often small, and different studies
298 Chapter Nine

use different methodologies, making it hard to pool their data. In order to correlate
the quality of hearing with the levels of noise players experience during performance,
it is first necessary to measure the intensity of the sound the orchestra creates. How-
ever, there has been no attempt to standardize the points at which sound intensity is
measured on the stage in different studies, nor the type or volume of music being per-
formed when the measurements are taken. The structure of the concert hall or recital
room will also be a significant factor. Ambient sound levels are higher if the player is
sitting close to a wall, which reflects the sound. Measurements of hearing sensitivity
in players are also not necessarily made at the same frequencies by different investiga-
tors. Further difficulties arise when it comes to interpreting the results obtained. Two
players of the same instrument and of similar age, sitting side by side in the orchestra,
may have quite different hearing thresholds, but it is difficult to determine whether
this is due to differences in their professional experience or to other factors. There are
many additional considerations that need to be taken into account in the analysis. For
example, unless the same group of musicians is tested at regular intervals, it is nec-
essary to find a control population for comparison, but who should be used? Should
it be those with the best hearing found within a population, that is, those who have
always lived in a quiet environment, or should the control group come from a noisier
urban environment more typical of that inhabited by the musicians? Then, what level
of loss should be regarded as indicative of impairment? For some clinicians, at least,
an increase in threshold of 20 dB at a given frequency is regarded as clinically signifi-
cant. This represents a tenfold loss of sensitivity, and though this may seem a major
impairment, it must be viewed in the context of the extreme sensitivity of the auditory
system. When measuring hearing thresholds, care must also be taken to allow some
time to elapse (usually forty minutes to one hour) after exposure to significant noise
levels before any auditory tests are carried out, as, if the noise is of sufficient inten-
sity, it may cause a temporary threshold shift. This lasts longer with increasing sound
intensity or length of exposure. As a consequence of these and other sources of vari-
ability, similar results from different studies have been used to justify quite opposite
conclusions. Nonetheless, the evidence that sound levels experienced during playing
can, over time, cause damage to hearing is accumulating. Sound intensities above 90
dB may by generated within the orchestra over a significant proportion of playing time
for certain pieces, and peak intensities in excess of 110 dB are sometimes experienced.
Off the concert platform, playing in small, bare practice rooms with a reverberant
acoustic may be flattering to the tone but will also maximize the intensity of the sound.
In many cases, the greatest sound intensity experienced by the musician comes from
his or her own instrument. Brass and upper woodwind players seem to produce the
highest sound levels. In the case of trumpet players, the increase in bore size of many
modern instruments is actually designed to raise the volume of sound produced. In-
terestingly, violinists often show an asymmetry in the hearing thresholds between the
two ears, the left ear being less sensitive, presumably because its greater proximity to
the belly of the instrument means that it experiences greater sound intensities.
Orchestras are now beginning to take sound exposure to their members seriously.
The Association of British Orchestras recently initiated a project entitled “A Sound
Ear” to look into the question. This has been made particularly timely by legislation
 Hearing and the Processing of Musical Sound by the Brain 299

from the European Union (directive 86/188) governing noise in the workplace, which
seeks to reduce exposure levels to a time-weighted average of 87 dB over a week of
five days of eight hours each, while the National Institute for Occupational Safety
and Health (U.S.) guidelines set the acceptable level at 85 dB. Every 3 dB increase
in sound intensity represents a doubling of the energy it contains and a halving of the
acceptable exposure time. So taking 87 dB as the starting value, exposure at 90 dB
should be limited to four hours, at 93 dB to two hours, and so on. Consequently, if
you are attending a rock concert where the sound reaches 110 dB, you have less than
two minutes before you exceed the recommended exposure time!
How acceptable levels can be maintained in orchestras performing with only acous-
tic instruments is less clear, both to the music industry and the legislators. A further
related EU directive (2003/10/EC) states that “the particular characteristics of the mu-
sic and entertainment sectors require practical guidance for the effective application
of the provisions laid down by this directive” and goes on to say that there will be a
transitional period to work this out. One way in which individual players (and indeed
concertgoers) can reduce the intensity of sound they perceive is to use earplugs. The
crude variety that simply block the external ear canal do not offer a practical solu-
tion because they do not reduce all frequencies to the same extent. High frequencies
are blocked most effectively, so the sound reaching the cochlea appears muffled and
distorted and does not accurately represent the ambient sound. This is obviously not
satisfactory in a musical context. Earplugs specially tailored for the needs of musi-
cians have been designed to reduce most frequencies in the audible spectrum to the
same degree, so wearing them is equivalent to turning down the volume on a stereo.
The best of the earplugs have to be custom-made to fit the ear canal of the individual
wearer and so they are not cheap. On the other hand, given the importance of protect-
ing the ears for musicians, this is probably a price worth paying. They can be designed
to reduce the sound intensity to differing degrees up to about 25 dB. In some instru-
ments such as the clarinet, there is significant conduction of sound from the instrument
to the ear through the bone of the skull due to the contact between the teeth and the
mouthpiece. In such circumstances, the wearing of earplugs may lead the player to
experience a buzzing noise due to bone conduction by this route, which may interfere
with their perception of the tone of the instrument (Hart et al. 1987).
The intensity of sound on the stage or in the pit obviously depends to some extent
on the proximity of other instruments, particularly the brass. One measure used to
reduce sound exposure to other players is to deploy screens behind the chair to create
a sound shadow. These are in use by some orchestras; however, they are only effec-
tive if they are sufficiently close to the head (about 15–20 centimeters) and lose their
efficacy if, for example, the player sits forward in the seat (Wright Reid 1999). Stage
layout can also be used to shield players from the sound of others. Moving players
farther apart will help, as does raising loud instruments like the brass above the rank of
seats in front, though the difference in height required to make a significant difference
(0.5–1.0 meters) may not always be achievable. In the pit, of course, the situation is
worse as players are generally closer together, all on the same level and in an enclosed
space whose walls may reflect the sound. Clearly varying the repertoire so that par-
ticularly loud pieces are mixed with quieter ones in performance and rehearsal also
300 Chapter Nine

helps to avoid discomfort among players. In addition, the conductor needs to be aware
of the state of his or her hearing. If an older conductor has a decreased sensitivity to
high frequencies and is unaware of it, he or she may ask for more from instruments
playing in this range. The result is not only potentially damaging for the hearing of
the players, but will of course not produce for the audience the balance of sound being
sought by the conductor.

Beethoven’s Deafness
The most celebrated case of deafness in musical history is, of course, that of
Beethoven. Remarkably, his body not only underwent a postmortem, but a copy of the
report is still in existence. In a rather macabre twist, although his ear ossicles and the
region of the skull containing the inner ear were retained for further examination, they
were later “lost or stolen”! The autopsy clearly supports his own accounts of serious
gastrointestinal problems including chronic diarrhea, vomiting, colic, and jaundice.
The immediate cause of his death appears to have been liver failure. This may have
been associated with inflammatory bowel disease, a condition that could explain
many of his other symptoms. The question of the origin of Beethoven’s deafness is
rather less clear. He describes its onset as being accompanied by continual whistling
and buzzing (i.e., tinnitus) and of first losing the ability to hear high frequencies. The
tinnitus started in about 1796 at the age of twenty-six, and he had lost much of his
hearing by 1801. In cases of major historical figures, there is always a tendency to
read too much into the little contemporary information still in existence. From a mu-
sical standpoint, it is the distressing nature of the symptoms that is most significant.
One conservative suggestion is that the deterioration in his hearing may have been a
combination of nerve deafness and otosclerosis probably unrelated to his other health
problems (Kubba and Young 1996).

HEARING AND THE BRAIN

How the Nervous System Codes Sound

The nerve cells that send auditory signals to the brain lie in the bony core of the co-
chlea. Like most other nerve cells, these carry information as pulses of electrical activ-
ity called action potentials. This is a digital signaling system and so it is very robust
and resistant to degradation as it transmits information through numerous relay sta-
tions, all the way up to the cortex. Two major strategies are used by the auditory sys-
tem to code sound frequency. The first is action potential frequency, and the second is
tonotopy (i.e., the relative position of the lines carrying signals representing different
frequencies). Even under the most favorable circumstances, action potentials cannot
be generated at frequencies above about 1,000 Hz and so they cannot code directly the
sound frequency across much of the auditory range. On the other hand, the tonotopic
map on the basilar membrane is unable to distinguish between sounds with frequen-
cies at the lower end of the range. The two systems are therefore complementary. For
the lower octaves of the auditory range, action potential frequency in single nerve cells
Figure 9-10. This figure shows two ways in which nerve cells of the auditory system
code sound frequency. A. At low frequencies, single nerve cells can accurately convey
frequency information. Each nerve cell can fire a digital signal (an action potential) at
the same point on each cycle of the sound wave; that is, the signal is phase-locked. B.
From 1,000 to 4,000 Hz, single nerve cells cannot fire action potentials fast enough
to keep up; however, they still fire phase-locked potentials on some cycles. By pooling
the information from many nerve cells supplying the same region of the basilar mem-
brane, the brain can still determine the pitch of the sound from the combined action
potential frequency. C. Above 4,000 Hz, the nerve cells still fire occasional action po-
tentials, but these are no longer phase-locked. Under these circumstances, the sound
frequency can only be determined by knowing the region of the basilar membrane
from which the signal originates (i.e., by using the principle of tonotopy).
302 Chapter Nine

matches sound frequency (fig. 9-10A), as long as each nerve cell fires an action poten-
tial in phase with the sound pressure oscillation. These phase-locked action potentials
are evenly spaced in time, and the sound frequency is a function of the time interval
between them. The upper theoretical limit for this type of signaling represents funda-
mental frequencies of notes of around C6 (two octaves above middle C). However,
even for lower notes, we need information about their harmonic structure in order to
appreciate the timbre of the sound, which gives each instrument or voice its distinc-
tive character. Between about 1,000 Hz and 4,000 Hz, it is still possible to use action
potential frequency as an indicator of the pitch of the fundamental if information from
many nerve cells that are stimulated by the same region of the basilar membrane is
pooled. Each nerve cell need not fire an action potential for every oscillation, but as
long as they are all phase-locked to the same place on the waveform, their collective
output can still be used to determine frequency directly (fig. 9-10B). Above 4 kHz, the
nerve cells lose their ability to phase-lock and instead fire randomly when stimulated
(fig. 9-10C). In this range, frequency can only be coded by tonotopy—the position
on the basilar membrane from which the signal is coming. It is therefore vital that
this order is retained within the bundle of axons that carry the signal to the brain and
that it is maintained through the many relay stations on the way. How this is ensured
is unknown, but the axon bundle forms a ribbon with the lowest-frequency fibers at
one edge and highest-frequency fibers at the other. Though it is possible to detect fre-
quency tuning above 5 kHz in nerve cells in various parts of the auditory system, we
cannot subjectively attribute pitch to the sound, so this change from frequency coding
to tonotopy appears to have important perceptual consequences. The highest note on
the piano (C8 at 4,186 Hz) is close to this boundary where tonotopy takes over (fig.
9-2). In addition to frequency, the intensity of the sound must also be conveyed to
the brain. This is encoded both by the number of action potentials generated and the
number of nerve cells activated, that is, the total number of action potentials generated
in a particular frequency channel of the auditory pathway.
The auditory information from the cochlea runs to the brain along the vestibulo-
cochlear nerve (the eighth cranial nerve) and passes through many relays within the
brain stem on the way to the auditory cortex (fig. 9-11). As it is passed from one to
another, the way in which the sound is coded changes. One of these relay stations is
the superior olivary nucleus, which we have encountered before when discussing its
role in the feedback control of hair cell activity. It is also involved in the signal pro-
cessing that makes it possible to identify the position of a sound source in space (see
below). Another important relay station lies in the roof of the midbrain (the inferior
colliculus). Here there is a map of auditory space that is linked to a similar map of
visual space in a nearby region (the superior colliculus). When we suddenly become
aware of a new sound, we may turn quickly toward it to look to the place from which
it appears to be coming. This is a hardwired reflex response that is controlled by
these midbrain auditory centers. The maximum frequency of the action potentials in
individual nerve cells is less in the inferior colliculus and cortex than in nerve cells
that leave the cochlea, so even coding for relatively low sound frequencies becomes
tonotopic at these upper levels (Fujioka et al. 2003).
Figure 9-11. The pathway by which information from the cochlea is carried to the auditory
cortex. The different shaded lines represent nerve fibers coding different sound frequencies. The
tonotopic order of these lines must be preserved throughout the entire pathway if the auditory
cortex is to decipher the frequency carried by each. By comparing the information reaching the
two ears, nerve cells in the superior olivary nucleus determine the direction from which a sound
is coming. With this information the inferior colliculi can create a map of the soundscape. The
superior olivary nucleus also feeds back onto the outer hair cells to control the sensitivity of the
cochlear amplifier.
304 Chapter Nine

Pitch and Missing Fundamentals

When presented with a complex sound, whether derived from a single instrument, a
number of instruments playing a chord, or from something more mundane like a car
engine, we can usually attribute to it a single pitch. This impression is created by an
analysis of the harmonic structure of the sound by the brain, and the pitch attributed
to the sound is generally that of its fundamental frequency. With practice it is pos-
sible to resolve some of the harmonics separately, but overriding the pitch perception
mechanism requires an effort of will and is a skill that must be consciously developed
by musicians. Combining sound from different sources to create the impression of a
single pitch is important in pipe organs. A given note can be generated by an open pipe
or by a stopped pipe (one whose open end is blocked) of half the length. For reasons
of cost and design, using shorter pipes is often more convenient, especially for low
stops. One problem with stopped pipes is that the sound generated contains only the
odd harmonics of the fundamental; the even ones are missing. As a result, the Tartini
tone generated by the odd harmonics has the pitch of the second harmonic (i.e., an
octave above the fundamental). However, if an open pipe with a pitch an octave higher
is played simultaneously, it provides not only the missing harmonics but also Tartini
tones, which support the fundamental frequency, so that together the two stops simu-
late an open pipe at the same pitch as the stopped one.
The mental process that allows us to attribute pitch to a complex sound can some-
times be fooled into creating the sensation of a fundamental frequency that is not
actually present in the sound spectrum—the so-called missing fundamental illusion.
For example, if a series of pure tones at 300, 400, 500, 600, and 700 Hz are presented
simultaneously, the perceived pitch is 100 Hz. This is logical, as these frequencies are
equivalent to the lower partials (excluding the second harmonic at 200 Hz) of a note at
100 Hz (sound presentation 9-3A). The perceived pitch is most strongly influenced by
frequencies that fall into what is known as the dominance region of our hearing range
where the ear is most sensitive, that is, between about 500 and 2,000 Hz (C5–C7). At
higher frequencies the missing fundamental illusion fails. When a series of tones at
1,000, 2,000, 3,000, 4,000, 5,000, and 6,000 Hz are presented together, the combined
pitch is perceived as 1,000 Hz, but if the lowest two harmonics are dropped, the pitch
becomes dominated by that of the lowest remaining tone at 3,000 Hz (sound presenta-
tion 9-3B).
So what is the mechanism of the missing fundamental illusion? If we return to the
example of the pure tones presented at 300, 400, 500, 600, and 700 Hz, one way in
which the pitch of 100 Hz could be derived by the nervous system would be from the
presence of the regular periodicity generated in the complex waveform produced when
these frequencies are combined, as this exhibits regular fluctuations at 10-millisecond
intervals. This is the cycle interval (or period) of the 100 Hz missing fundamental. We
have seen already that the nervous system can code frequency information by summing
activity in different nerve cells all stimulated by the same part of the cochlea. An exten-
sion of this type of analysis could generate the perceived frequency here (Yost 2000).
However, this is not the whole story. If we now consider the series 850, 1,050, 1,250,
1,450, and 1,650 Hz, this has a periodicity of 200 Hz even though all of the notes are
also high harmonics of 50 Hz. The perceived pitch appears close to, but not exactly
 Hearing and the Processing of Musical Sound by the Brain 305

200 Hz (sound presentation 9-3C). Instead it lies at 208.3 Hz, which represents the
fundamental pitch whose harmonics would most closely match the tones presented; in
other words, it represents a “best fit” solution. Other studies, however, have cast doubt
on a mechanism of pitch perception based purely on periodicity. If, instead of a small
number of pure tones, the sound presented has many intermediate frequencies of vary-
ing intensity across the same range, but which tend to be of greatest amplitude at 500,
750, 1,000, 1,250, 1,500, and 1,750 Hz, the periodicity of the combined waveform will
be completely lost, yet the perceived pitch remains at 250 Hz (Yost 2000).
In view of what has been said so far about the role of tonotopy in coding the fre-
quency of a pure sound, it is perhaps surprising that recent studies have suggested
that it can contribute to the perception of pitch (Penagos, Melcher, and Oxenham
2004). It appears to be particularly significant for the low-frequency components of
complex sound because, as the higher harmonics become closer and closer together,
the corresponding peaks of the traveling waves on the basilar membrane cannot be
clearly distinguished from one another. As the individual waves interfere, the resulting
complex does show a periodicity related to the fundamental frequency, though in itself
this does not appear to make a major contribution to pitch recognition. The origin of
pitch perception is clearly an area that requires further investigation. However it is
achieved, the perception of pitch by the brain requires a sophisticated mechanism for
pattern recognition. For notes of low pitch, the illusion of the missing fundamental
can be created with as little as two pure tones representing its harmonics. It can even
be achieved when the two tones are presented to different ears. This demonstrates un-
equivocally that the illusion is created not in the ear, but in the central nervous system
(Rasch and Plomp 1999).
The obvious question that arises from our consideration of pitch is what is the sig-
nificance for hearing in general, and music in particular, of the way in which complex
sounds are analyzed by the brain? If each sound frequency was treated separately by
the brain, the soundscape would be incredibly complex and difficult to interpret. The
ability to attribute a single pitch (and therefore source) to harmonically related clusters
of frequencies greatly simplifies the auditory world and allows us to interpret it in a
logical way. The relative strengths of the harmonics in complex sounds gives them
a distinctive timbre, which further helps us to identify their source. From a musical
point of view, a group of instruments playing in an ensemble can present an impres-
sion of a sound of a single pitch when they contribute to a chord (albeit colored by the
particular harmonics they contribute). However, the flexibility of the analysis carried
out by the nervous system allows this to be realized without prejudicing our ability
at other times to follow several melodic lines created simultaneously by the same
instruments. In addition, the illusion of the missing fundamental is vitally important
in a number of circumstances. As with difference tones, the devious minds of organ
builders have again found that it can contribute to the Acoustic Bass stop. Indeed, for
some notes in this stop, a low difference tone may also represent a missing fundamen-
tal and so the cochlear and the mental illusion reinforce one another. However, the
missing fundamental illusion has a much broader significance for instruments with a
low frequency range in their register. A reexamination of figure 9-2 will remind you
that the threshold for hearing varies considerably with frequency. As sound frequency
306 Chapter Nine

descends from 100 to 30 Hz, this rises from about 40 to 80 dB. As this frequency
range represents the bottom two octaves of the piano (A0–G2), we are not dealing
with unfamiliar territory. For these notes, eliminating the fundamental frequency and
the next couple of harmonics makes almost no difference to the perceived pitch or to
the timbre. The implication is that the perception of pitch in this range is almost en-
tirely dependent on the missing fundamental illusion, particularly if the notes are not
loud. It also explains why the acoustic bass can work so well on the organ. The low-
est notes on the organ descend even farther than those of the piano. A thirty-two-foot
open pipe produces a fundamental frequency of only 16 Hz, which is actually below
the threshold for complex sounds, lying as it does at around 30 Hz for human hearing
(Pressnitzer, Patterson, and Krumbholz 2001). In the small number of acoustic organs
that have this stop, the bottom few notes on the thirty-two-foot rank may actually all
play the same pipe (Bicknell 1999)! This gives a suitably gratifying rumble of sub-
jectively indeterminate pitch, as the harmonic structure of the notes generated by the
wooden pipes in this rank is weak. The perceived pitch can then be dictated by other
stops representing the upper harmonics for the required notes. The missing fundamen-
tal illusion is also important in an orchestral context. The average spectral peak of the
sound produced by the orchestra lies at about 500 Hz (C5), from which it rapidly de-
clines. For instrumental or vocal soloists, and indeed for solo passages from orchestral
players, the missing fundamental illusion will make a significant contribution to our
ability to follow their melodic line against a loud orchestral accompaniment.
At the lowest end of the range of the pipe organ, the appreciation of sound at or be-
low the auditory range may be accentuated by the effects of vibration on sensory end-
ings throughout the body and not just in the ear. When of reasonable large amplitude,
the low-frequency vibrations generated within the organ will be conducted through
the walls and floor of the building as well as through the air. These can be detected by
vibration sensors in the skin and joints, or even those associated with structures inside
the chest or abdomen. Obviously such sensations can really only be experienced in a
live venue such as a church or concert hall and are hard to reproduce in recording, no
matter how tolerant the neighbors may be!

Analysis of Sound by the Brain

The auditory pathways that originate in the cochlea finally terminate in the region of
the forebrain called the primary auditory cortex. This lies in the temporal lobe and
forms the central two-thirds of a ridge on its upper surface known as Heschl’s gyrus
(fig. 9-12). Within the primary auditory cortex, the frequency of sound is mapped
in an orderly tonotopic fashion. From here this raw auditory information is sent out
into a belt of secondary auditory cortex that includes the region known as the planum
temporale, which lies behind and above Heschl’s gyrus. It has been proposed that
information on pitch and timing of sound is processed here before being distributed
to other areas of the temporal and frontal lobes for higher-level analysis (Griffiths
2001; Griffiths and Warren 2002). The processing of different aspects of auditory
information is modular. Analysis of the temporal aspects of sound, such as meter and
rhythm, and melodic aspects involving pitch contour, phrasing, and harmony are to a
Figure 9-12. Auditory cortex. A. The primary auditory cortex lies on the upper edge of the tempo-
ral lobe. This is surrounded by the secondary auditory cortex, where higher-level auditory process-
ing takes place. One region of the secondary auditory cortex is Wernicke’s area (left hemisphere
only), which is involved in language processing. B. A horizontal section through the brain along the
plane marked by the dashed line in A. This reveals more of the upper surface of the temporal lobe,
which is obscured by the overhanging regions of the frontal and parietal cortices. The primary
auditory cortex takes up much of a region known as Heschl’s gyrus. Behind this lies the planum
temporale of which Wernicke’s area forms a part. C. This shows a schematic representation of the
sound frequency maps in the primary auditory cortex; however, it is now thought that the primary
and secondary auditory cortices may contain several distinct maps of this type.
308 Chapter Nine

considerable degree carried out in different places. This can be dramatically revealed
following damage to discrete areas of the cortex. For example, a patient may be able
to discriminate between different tones in simple tests, yet not be able to distinguish
tonal and atonal music. The principle of modular processing of different parameters of
sensory information is well established in other systems. In vision, for example, color,
movement, and form are processed in different areas of the cortex.
The primary auditory cortex, like the cortex in general, is a thin sheet of gray mat-
ter made up of an array of columns that run vertically through the sheet. Within each
column the nerve cells respond to the same type of information. In many regions of
the primary auditory cortex, each column responds to a particular frequency. A com-
monly presented but rather simplistic view of its organization is that the columns are
arranged in an orderly fashion to form a single tonotopic frequency map, with those
toward the anterior and lateral edge responding to low frequencies and those lying
more posteriorly and medially, to high frequencies (fig. 9-12C). In fact there appear
to be several such maps (perhaps as many as seven) within the primary and adjacent
secondary areas of the auditory cortex (Kaas, Hackett, and Tramo 1999; Talavage et
al. 2004). Information on the loudness of the sound is also coded here, as some of the
nerve cells respond best to their preferred frequency when it is at a particular intensity.
In addition, some rows of columns are excited most strongly by sound that comes
from both ears, while others respond best to sound from one ear and are inhibited
when it comes from both (Schreiner and Winer 2007). There has been considerable
debate as to whether the primary auditory cortex responds predominantly to individual
sound frequencies or to the overall pitch of complex sounds, or indeed both (Pierce
1999). Studies that record the electrical activity of single neurons often suggest that
primary auditory maps are based on frequency, whereas those that monitor the si-
multaneous activity of many neurons suggest that these assemblies code for pitch.
However, it has recently been demonstrated that some auditory neurons respond not
only to a pure tone of a given frequency but also to a harmonic series that represents
the same frequency as a missing fundamental. They do not, however, respond to the
individual pure tones that make up such a series (Bendor and Wang 2005) and so they
can be regarded as being pitch sensitive. Evidence is beginning to emerge that while
the tonotopic maps in the primary auditory cortex are based on frequency, neurons that
respond to pitch are present in the adjacent areas of the secondary cortex that lie more
anteriorly or laterally on the upper surface of the temporal lobe (Bendor and Wang
2005, 2006; Penagos, Melcher, and Oxenham 2004).
Neurons in the primary auditory cortex carry information not only on frequency but
also on the temporal aspects of the sound. This is processed in the secondary auditory
cortex and elsewhere in the brain to provide information on rhythm and meter (the pat-
tern of stresses in a rhythmic beat that allows us to distinguish a waltz from a march).
Both the left and right primary auditory cortices respond to musical sounds, but while
there appears to be a right-sided bias for frequency information, the temporal com-
ponent is more bilaterally represented (Evers et al. 1999). Indeed, some studies even
suggest slightly greater activity on the left side for this type of information (Jamison
et al. 2006), though, like many aspects of the analysis of auditory processing, this
remains a matter of debate (Zatorre, Chen, and Penhune 2007).
 Hearing and the Processing of Musical Sound by the Brain 309

The development of modern imaging techniques that enables us to measure vari-

ous aspects of neuronal activity in the intact living brain has led to a proliferation of
studies into how musical sound and language are analyzed; however, many aspects
of this remain highly controversial. This stems partly from the difficulty of design-
ing experiments that effectively isolate particular parameters of musical sound. For
example, without rhythm, a melody loses much of its character to become simply
a series of intervals, and this may not be analyzed in the same way or in the same
region of the brain as an intact melody. Rhythm is easier to isolate in the absence of
pitch or melody, but is closely linked to meter, which may be seen as a higher-order
property of rhythm analysis, interpreted by a different network. Furthermore, though
the new technology provides unprecedented opportunities, the complexity of the brain
circuitry makes sensible interpretation of data very challenging. A single region may
be involved in a range of tasks, and establishing the common denominator can be
extremely difficult. We have seen from the previous chapter the many other problems
that beset the unwary when carrying out such studies, but these methods nonetheless
offer unprecedented opportunities for studying this area.
One theory of the way that auditory processing is carried out in the cortex postulates
that this is done along two processing streams. One reason for this notion is that a sim-
ple way of explaining how the visual system operates is based on a similar idea, which
suggests that there is a ventral stream concerned with processing information about
object identity (what) and a dorsal stream concerned with object position (where). The
first auditory stream is said to run anteriorly along the upper and lateral surfaces of the
temporal lobe, through the secondary auditory regions including the planum polare,
which lies toward the anterior end of the lobe (fig. 9-13A). This stream is also said to
encompass some areas in the ventral regions of the frontal lobes. It is proposed that,
like the ventral stream in the visual system, this is concerned with identity (what). As
the equivalent visual stream also extends into the temporal lobe, this theory has the
advantage of providing a potential mechanism for bringing together the visual and
auditory aspects of object recognition. The proposed second stream runs dorsally and
medially through the parietal lobe and into the dorsal regions of the frontal lobe. Here
the analogy with the “where” stream of the visual system is rather less helpful. The
most obvious role of this stream is to link auditory and motor activity, and it appears
to underpin rhythm perception (fig. 9-13B). It has been suggested that the planum
temporale may form a central node for the distribution of information along the two
streams. One hypothesis is that by matching the harmonic structure and temporal
patterning of incoming complex sounds with previously remembered templates, the
planum temporale can then distribute the information appropriately to other cortical
areas for higher-level processing (Griffiths and Warren 2002). In this regard, it is
interesting to note that in European musicians the planum temporale has been shown
to be more active when hearing culturally familiar (Western) music than culturally
unfamiliar (Chinese) music (Nan et al. 2008). On the left side of the brain, the planum
temporale includes Wernicke’s area, which is crucial for linguistic comprehension,
and it has a strong projection to a region in the left frontal lobe called Broca’s area,
which is associated with verbal fluency and intonation. As we will see, Broca’s area
is also active when the brain is analyzing music.
Figure 9-13. Auditory processing streams in the brain. A. The ventral stream is
thought to process pitch contour, melody, and timbre. Auditory information runs
from the primary auditory cortex to the planum temporale, then along the upper
surface of the temporal lobe to the planum polare. A tributary of this stream runs
into the prefrontal lobes and may be concerned with melodic memory, among other
things. B. The dorsal stream is strongly implicated in processing rhythm. This recruits
the motor network of the brain, including areas of the cortex involved in movement
control as well as the basal ganglia and cerebellum.
 Hearing and the Processing of Musical Sound by the Brain 311

The recognition of melody is based on the identification of pitch changes over

time. There is accumulating evidence that this involves the posterior regions of the
secondary auditory cortex, including the planum temporale (Patterson et al. 2002;
Liegeois-Chauvel et al. 1998; Peretz and Zatorre 2005). As the analysis extends to the
perception of melody, activity appears to spread along the superior temporal gyrus
to the planum polare (fig. 9-13A). This therefore suggests a hierarchical sequence
of processing along the ventral stream. Several distinct processes may be involved.
Determination of the overall contour of the melody (disregarding interval size), may
use a mechanism that would also be useful for analyzing speech inflection. Musical
analysis would also require matching pitch relations regardless of absolute frequency
(i.e., recognizing the similarities between transposed melodies) or key violations of
individual notes in the sequence. Most of the tasks used to obtain these results involve
memory to compare whether two presented melodies are the same or different. As a
result, regions of the brain implicated in associative and working memory are also
implicated, for example, the hippocampus (temporal lobes) and inferior frontal lobes
(Peretz and Zatorre 2005; Watanabe, Yagishita, and Kikyo 2008). Given that timbre
is likely to be a significant component in sound source recognition, one might expect
that it would also be processed in the ventral processing stream. Cortical responses
related to timbre have so far received relatively little attention, but both its perception
and its mental imaging lead to activity in and around the superior temporal gyrus and
on the border between Heschl’s gyrus and the planum temporale (Halpern et al. 2004;
Menon et al. 2002). Attending to harmonic discrepancies activates similar regions on
the ventral stream, down into the planum polare (particularly in the left temporal lobe)
in addition to the frontal cortex in the region of Broca’s area and its equivalent on the
right side (Passynkova, Sander, and Scheich 2005; Parsons 2001). A similar pattern
(with the additional involvement of motor areas) was noted during vocal repetition
of a melody or a vocal harmonization task, though in this study the activations were
bilateral (Brown et al. 2004).
When we listen to rhythmic music we often find our feet tapping or our head nod-
ding to the beat or may experience a strong urge to dance. If this observation appears
self-evident, compare it to the complete lack of response we experience when watching
rhythmic movement in the absence of sound. It is only through the auditory system that
the urge for movement is generated. It will therefore come as no surprise to learn that
the motor areas of the brain play a central role, not only in the production of rhythmic
movement, but also in the perception of auditory rhythms (fig. 9-13A). Our response
to rhythm does not rely simply on an interpretation of incoming signals, but also on an
ability to predict how the rhythm will continue. Without this, our dancing, foot tapping,
and indeed music making would lag behind the beat. This implicates the brain circuitry
involved in the temporal patterning of sequential movements and is consistent with the
observation of activity in the basal ganglia, the cerebellum, and the supplementary mo-
tor cortex in response to auditory rhythms (Limb et al. 2006; Grahn and Brett 2007).
This is present even in the absence of overt movement, during which the primary motor
cortex remains silent. It has been proposed that while the basal ganglia and supplemen-
tary motor area play a key role in the timing of motor activity over periods of several
seconds, the cerebellum is responsible for the precise synchronization movement in
312 Chapter Nine

the subsecond range (Zatorre, Chen, and Penhune 2007). In basal ganglion diseases
such as parkinsonism, in which motor sequences break down, leading to a slowing or
freezing of movement, music or the beat of a metronome can be an effective means of
starting it up again and can help the individual with the condition to maintain a regular
walking cadence (Suteerawattananon et al. 2004; Zatorre, Chen, and Penhune 2007;
Sacks 2007). Rhythm also evokes activity within the frontal premotor cortex including
Broca’s area, which lies in its ventral region. However, as the rhythms (either perceived
or produced) become progressively more complex, the dorsal parts of the premotor
cortex are increasingly recruited (Zatorre, Chen, and Penhune 2007). The basic pattern
of activity described so far appears equally applicable to both musicians and nonmusi-
cians, though musicians show additional foci of activity in the prefrontal cortex, which
may indicate a greater involvement of working memory (the neural equivalent of com-
puter RAM) in the analysis of the temporal structure of rhythm (Chen, Penhune, and
Zatorre 2008). The perception of meter appears to have a different origin, requiring the
involvement of the anterior superior temporal gyrus (the upper surface of the temporal
lobe some distance in front of the primary auditory cortex) (Grahn and Brett 2007;
Liegeois-Chauvel et al. 1998).
The history of our understanding of the role of Broca’s area in speech and in the
analysis of musical sound is instructive in demonstrating how the properties of a given
brain region may fit it for a range of different functions. It should also be seen as a
warning against applying simplistic functional labels to particular regions of the brain,
something that is all too tempting with functional imaging. Broca’s area lies on the left
inferior region of the frontal lobe and was first described by Pierre Paul Broca in 1861,
following examination of the brains of two of his patients who had lost the ability to
speak. The brains are still in existence and have recently been scanned to reveal that the
lesions extend considerably beyond what is now known as Broca’s area (Dronkers et al.
2007). In the 1990s, animal experiments revealed that Broca’s area and its equivalent on
the right side of the brain contain what are known as “mirror neurons.” These respond
both when we perform an action and when we watch the same action being performed
by another. It was subsequently discovered that some mirror neurons also respond to the
sounds that are generated during such actions (Kohler et al. 2002). Though for ethical
reasons the properties of single neurons in Broca’s area cannot be studied directly in
humans, circumstantial evidence strongly supports the likelihood that some mirror neu-
rons within our own brains also have this property (Gazzola, Aziz-Zadeh, and Keysers
2006). This may help to explain the long-established association of Broca’s area with
speech (Zatorre, Chen, and Penhune 2007; Gazzola, Aziz-Zadeh, and Keysers 2006).
Neurons in Broca’s area respond when sounds are created using movements of the hand
or the mouth, and the level of activity has been shown to be higher in musicians than in
nonmusicians (Bangert et al. 2006). A similar response is seen in the equivalent region
on the right side of the brain, though it is not as intense. Though it doesn’t itself contain
mirror neurons, the secondary auditory cortex in the middle temporal gyrus may also
be involved in this system. Functional imaging studies now suggest that Broca’s area
contributes to the processing of several aspects of musical syntax including harmonic
structure, melody, rhythm, and timbre (Brown and Martinez 2007; Koelsch 2006),
though its precise role in these activities has yet to be explained.
 Hearing and the Processing of Musical Sound by the Brain 313

Consonance and Dissonance

Certain combinations of pure or complex tones are perceived as more harmonious than
others. In general pairs of notes, those whose fundamentals are related by simple ratios
appear to us more consonant than those with more complex ratios. Examples of con-
sonant intervals are the octave (ratio 2:1), the perfect fifth (2:3), and the perfect fourth
(4:3), while the minor second (16:15) and the tritone (45:32) are dissonant intervals
(Tramo et al. 2001). Several explanations have been put forward for this. One is that
consonant combinations of complex sounds tend to have harmonics that are common
or are numerically related. For the perfect fifth, for example, half of the harmonics
of the upper note coincide with harmonics of the lower one (Burns 1999). This is re-
flected in a regular periodicity of the combined waveform and in the response of the
population of neurons in the auditory nerve that respond to the sound thus generated.
It is the same phenomenon discussed above in the context of the perception of missing
fundamentals, and indeed these will be generated by consonant note combinations.
For dissonant intervals, not only are any periodicities much weaker, but they would
be indicative of fundamentals that are not related to the notes being played (Tramo et
al. 2001). A second theory of consonance is that the circuitry of the auditory system
is designed to look for harmonic relationships. Again this is the system that assigns a
single pitch to complex sounds. Any mismatch of the harmonic components that does
not allow the fitting of a pitch to the set of frequencies presented is therefore very
noticeable. At the very least, the perceived pitch clashes with some of the harmonics
so that they stand out as noticeably alien components. When subjects are presented
with a series of chords that include unexpected dissonances, activity is seen in the
region in front of the primary auditory cortex (the anterior superior temporal gyrus),
the planum temporale, and the region of the frontal lobe in and around Broca’s area
and its equivalent on the right hemisphere (Koelsch 2006; Maess et al. 2001; Parsons
2001; Passynkova, Sander, and Scheich 2005).

Cortical Plasticity in the Auditory System

From previous chapters on the sensory and motor areas of the cortex, the idea that the
brain is molded by experience will be a familiar one. This is equally true of the audi-
tory cortex. In a study of Heschl’s gyrus using brain imaging techniques, the region
occupied by the primary auditory cortex in amateur and professional musicians was
found to be more than twice as large than that in nonmusicians (Schneider et al. 2002;
Gaser and Schlaug 2003; Luders et al. 2004). In professional musicians it is also larger
than in amateurs. The auditory cortex of musicians shows a greater response to the
sound of their own instrument than to the sound of other instruments (both of which
are complex sounds composed of a fundamental frequency and many harmonics), as
well as to pure tones of a single frequency. There is a general trend for the magnitude
of the signal generated to the sound of the player’s own instrument to be greatest in
those who started playing before the age of about eight (Pantev et al. 1998, 2003).
The left and right side of the brain are not symmetrical in terms of function. For ex-
ample, linguistic skills are usually located mainly on the left side, close to the auditory
314 Chapter Nine

areas. The auditory cortex itself also shows asymmetry. It appears from some studies
that perception of pitch is primarily located on the right side but that the temporal pat-
tern of the sound is analyzed mainly on the left side (Liegeois-Chauvel et al. 2001;
Zatorre 2001). It has also been suggested that a right-sided bias for analysis of harmony
and melody may be greater in females (Evers et al. 1999). It has been proposed that the
reason that analysis of language lies on the left is that it is mainly the temporal changes
in the sound that are important for interpreting speech. On a slightly longer time scale,
the inflections found in speech that carry nonverbal grammatical information have some
similarities with melodic phrases. When amateur musicians and nonmusicians are set
difficult tasks requiring analysis of the harmonic and melodic structure of a piece of
music, the regions of the brain involved in interpreting music on both sides of the brain
are used in relatively equal measure. In highly trained professional musicians, there is
greater activity on the left side of the brain than on the right, implying that the infor-
mation is being processed in a different way (Altenmuller 2001; Ohnishi et al. 2001;
Evers et al. 1999). This is probably more marked in those whose formal musical train-
ing started early, as the brain is more able to alter its internal connections in childhood.
The types of changes seen here are not unique to musicians. It is known, for example,
that in those who are brought up to be bilingual from birth, the region of the cortex that
controls the muscles used in speech is identical for both languages; however, when a
second, fluently spoken language is learned later in life, the areas controlling speech in
the two languages are separate (Kim et al. 1997).

Perfect Pitch
Pitch and interval discrimination are important abilities that musicians must develop.
The majority have what is called good relative pitch, which is to say that when pre-
sented with a note, they can sing on demand another that is separated from it by a
particular musical interval (e.g., a minor third or a fifth), or on hearing a named note,
they can name the pitch of another that is played soon after. This is based on short-
term memory of the first note, which can be erased with a burst of white noise, a
silent interval of a minute or less, or a brief distracting acoustic or nonacoustic task.
Perfect or absolute pitch is less common and is the ability to name the precise pitch of
a single note played in the absence of any reference cues. An individual with perfect
pitch can therefore typically name up to seventy notes or “varieties of pitch,” whereas
most ordinary mortals can recognize only six to eight ranges of pitch without refer-
ence points (Zatorre 2001, 2003). It should be noted, however, that those with perfect
pitch are not more accurate at identifying pitch deviations than those without absolute
pitch; their unusual abilities are confined to pitch labeling only and do not extend
to pitch perception (Levitin and Rogers 2005). The problem of pitch recognition is
similar to that posed by colors. Our computer monitor may be capable of producing
several million shades, but we can accurately label little more than the seven spectral
colors we define in the rainbow. We have all had futile arguments over such common
intermediate shades as turquoise, or had the embarrassment of guessing a paint shade
in a home improvement store, only to discover when we get the paint tin home that it
is nothing like the color on the wall. The limitations of the analogy are also instruc-
 Hearing and the Processing of Musical Sound by the Brain 315

tive. One of the requirements for developing an absolute hue perception comparable to
absolute pitch would be the presence of a system for naming discrete shade intervals
across the continuum of the visible spectrum. However, the musical intervals of the
scale have a greater internal structure, as they are formed from a repeated series of
steps related to the harmonics of a particular note. In modern tuning, the interval of
the octave is absolutely precise and even small deviations from this are easily detected
by the beats between two notes. The common property that we perceive in notes one
or more octaves apart is described by the word “chroma” but ironically, despite the
name, such a precise scale cannot be devised for color.
To acquire perfect pitch it is necessary both to have the innate ability and to be ex-
posed at an early age to an environment in which notes and their names are presented
simultaneously. Interestingly, and for reasons that are currently unknown, this appears
to be easier for the white notes on the piano than the black ones (Pierce 1999). Most
musicians with perfect pitch start their musical education before the age of seven
and during their early learning period have not had their absolute judgment of pitch
clouded by consciously learning relative pitch. Many adults have struggled long and
hard in an attempt to develop perfect pitch, with an almost total lack of success. What
we are seeing here is an example of a phenomenon well known in brain development
called a critical period. Though the brain remains able throughout life to change its
wiring patterns in response to the demands made upon it by changing conditions,
the degree to which this is possible declines with age. Many circuits can be radically
altered only during a short critical period of a few months or years in early life. For
example, if a child has a squint (a walleye), this should be corrected before the age
of six, otherwise the circuitry for binocular vision, which requires both eyes to point
in the same direction, will not develop. The ability to develop perfect pitch may also
depend on genetic factors (see below), as it is seen more often in children of the same
parents than would be expected by chance and is more common in some ethnic groups
than others, notably in people of Asian origin (Zatorre 2001). It does not, however,
appear to be due to the inheritance of a single gene.
Perfect pitch is associated with an increase in the asymmetry of the planum tempo-
rale, the area of association auditory cortex that lies behind Heschl’s gyrus. This was
demonstrated in a comparison of brain structure between thirty musicians, some of
whom had perfect pitch, and thirty nonmusicians. In the nonmusicians, and the mu-
sicians without perfect pitch, the left planum temporale was 22 to 26 percent larger
than the one on the right. However, the greatest asymmetry was seen in male musi-
cians with perfect pitch, in whom the left planum temporale was nearly 80 percent
larger than on the right side (Schlaug et al. 1995; Luders et al. 2004). This was not
due to its being larger in absolute terms, but because the right planum temporale was
actually smaller than normal in this group (Schlaug et al. 1995; Keenan et al. 2001).
In female musicians, by contrast, there was no increase in the size of the left planum
temporale associated with perfect pitch (Luders et al. 2004). Interestingly, studies of
brain function suggest that perfect pitch may be related to the absence of a process
that refreshes short-term working memory of pitch change. In those who only have
relative pitch, this process is required to establish the interval relationships between
different notes.
316 Chapter Nine

Perfect pitch is only absolutely “in tune” if the reference pitch the brain acquires
correlates with international concert pitch. It has been said of those who have it that
“perfect pitch” is the pitch of their mother’s piano, so parents who aspire to raise
musical geniuses should not skimp on paying for the tuner! While it may have many
advantages, perfect pitch also has its drawbacks. A solo singer who persists in main-
taining concert pitch when the choir goes flat may not be asked back. Much has also
been made of the fact that the reference pitch of individuals with perfect pitch may
not remain constant throughout life. Many (but not all) individuals with this ability
have reported that their perception of pitch changes with age, sometimes by several
half-steps in one direction or another (Pierce 1999). If the musician has definite feel-
ings about the character of different keys, this can be disturbing. Suggestions based
on anecdotal evidence that absolute pitch changes during pregnancy or the menstrual
cycle have not been supported by more objective testing; however, it does appear to
be true that in both males and females, pitch can drift up and down over a range of
10–20 Hz over a period of a few months.

Tone Deafness and Amusia

President Ulysses S. Grant once explained to a reporter that he knew two tunes: one
of them was “Yankee Doodle” and the other wasn’t. He suffered from tone deafness
(congenital amusia) and was not simply unmoved by music but found it positively ir-
ritating (Munte 2002). Tone deafness may be quite common; one survey of six hundred
students found an incidence of 10 percent (Shuter-Dyson 1999), but a more generally
accepted figure is that it affects 4–5 percent of the population. There is some evidence
that it may have a genetic component in at least a proportion of those affected (Peretz
and Hyde 2003). Those who are tone deaf find it impossible to recognize or hum a
melody, to sing in tune (or even to recognize that they are not doing so), regardless
of whether they have had musical training in childhood. It is found in people of all
mental abilities and does not preclude good verbal skills in one or more languages. In
many people with congenital amusia (though not all), the ability to perceive rhythm
is unaffected. Even small children can normally recognize dissonance, and the lack of
this ability in those who are tone deaf seems to reflect a difficulty in identifying small
deviations in pitch and in relating tones to musical scales (Peretz and Hyde 2003; Fox-
ton et al. 2004). Strangely, this difficulty is only experienced in the context of music.
The perception of inflection in language, such as the rising tone at the end of a question
in English and other Indo-European languages, is essential for interpreting the precise
meaning of spoken phrases. In languages such as Mandarin, inflection is even more
important in giving meaning to words. These tasks, however, pose no problem to those
who are tone deaf. The reason resides partly in the relative sizes of the pitch excursions
in language and music. While recognition of melodic lines requires identification of
pitch increments as small as one-twelfth or less of an octave, linguistic inflections typi-
cally extend over half an octave or more, so a deficit in pitch perception of sufficient
magnitude to interfere with the interpretation of music (typically one or more tones)
will generally have little consequence for the nonverbal elements of spoken language
 Hearing and the Processing of Musical Sound by the Brain 317

(Peretz 2002). However, that is not the whole story, as pitch changes in the range typi-
cal of speech are more easily detected when delivered as words than when not (Ayotte,
Peretz, and Hyde 2002). Comparisons of the brains of normal subjects and individu-
als with congenital amusia using magnetic resonance imaging analysis have revealed
anatomical anomalies associated with tone deafness. There is a thickening of the gray
matter in the right superior temporal gyrus and both thicker gray matter and reduced
white matter in the right inferior frontal gyrus (Hyde et al. 2006, 2007). This is thought
to be due to malformation of the cortex during development. The superior temporal gy-
rus is strongly implicated in the fine pitch discrimination and sequential pitch analysis
that underpins the perception of melody, while the inferior frontal gyrus is thought to
be involved in melodic pitch encoding and memory.
Musical disability that arises as the result of brain damage or disease is a quite
different phenomenon. This is usually referred to as acquired amusia, a name that
covers a number of disorders, not only of the central processes underlying music
recognition and interpretation but also of the reading, composition, and performance
of music. When it produces symptoms similar to tone deafness, the areas affected are
generally the primary and secondary auditory cortices on the superior temporal gyrus.
One composer who developed amusia of a different type in later life was Ravel. He
ceased to compose in his midfifties, and his amusia was a symptom of a degenerative
age-related disease that today would probably be labeled frontotemporal dementia
(Sacks 2007). Though he retained some memory of his own compositions, he became
unable to sight-read or name notes and also found it difficult to copy music either
from a manuscript or from dictation (Marin and Perry 1999). Oliver Sacks has even
suggested that the monotonous repetition and lack of development that characterizes
Bolero may have been an early indication of the onset of his condition.
The degree to which case studies of individuals who have suffered localized damage
to the cortex can dramatically illustrate how musical information is processed is well
demonstrated in a series of studies by Peretz (2002). She describes one young woman
who underwent several operations to repair ruptured blood vessels in the brain, which
resulted in a loss of part of the auditory cortex on both sides of the brain together with
regions of the adjacent frontal cortex on the right side. It is a testament to the ability
of the brain to compensate for damage that she not only retained her linguistic abilities
but also suffered no loss of intellect or general memory function; however, she was no
longer able to recognize or memorize music or sing a tune. Though not a musician, she
was brought up in a musical family and music had been an important part of her life
before surgery. She is described as suffering from amusia without aphasia (aphasia is
the loss of the ability to produce or understand speech). A mirror image of this deficit
was seen in the case of the Russian composer Shebalin. He suffered two strokes (loss
of blood supply to part of the brain), which affected a large area of the left half of
the cortex. Though he lost the ability to speak, during the remaining four years of his
life he nevertheless composed a symphony, four quartets, two sonatas, eleven songs,
and more than a dozen chorales. Shostakovich, who was a contemporary, considered
that this music was indistinguishable in quality from that composed before his illness.
Shebalin’s deficit was therefore aphasia without amusia.
318 Chapter Nine

Auditory Synesthesia
For a small number of people, hearing music generates strong involuntary sensations
normally associated with senses other than hearing. For example, particular notes or
keys may appear to assume specific colors. This is just one form of a wider phenom-
enon called synesthesia, in which stimulation of one sense produces illusions belong-
ing to another. In its other manifestations, numbers may be associated with colors, or
touching certain objects may induce the sensation of a particular taste. Though there
has been some controversy in the past about whether this is a real phenomenon and not
simply a learned response, or perhaps a drug-induced state or even a form of fakery, the
evidence has become increasingly cogent (Ramachandran and Hubbard 2003). The fact
that it appears to have a genetic component (i.e., it runs in families) helps to support
this view. Because it was apparently experienced by some well-known composers, syn-
esthesia triggered by music has some historical significance. Messiaen, Rimsky-Kor-
sakoff, and Scriabin were all apparently synesthetes. The latter two, who were friends,
could not agree on the colors associated with particular keys but with characteristic
robustness, Scriabin came up with a theory to explain Rimsky-Korsakoff’s “mistaken”
perceptions. Scriabin’s synesthesia led him to experiment with the color organ, an in-
strument that presented colored light as an accompaniment to music and for which he
wrote the orchestral piece Prometheus, the Poem of Fire.
The origin of synesthesia is unknown, though it is thought to be a product of the
way in which the brain is wired. When we think of a place, an object, an animal, or
a person, we bring together stored information from many of our senses to create a
holistic impression in our mind. Information from any one sense may trigger the same
complete memory. An important design principle of the brain is therefore to bring
different sensory streams together. The difference with synesthesia is that the associa-
tions either are with abstractions (e.g., linking tonality with color) or involve matching
sensations that cannot be generated by the same object. This suggests that its roots lie
in atypical connections between brain regions that deal with abstracted sensory infor-
mation. On theoretical grounds, a particular region of the cortex that lies above and
behind the planum temporale has been proposed as a site for synesthetic associations;
however, there is as yet no direct evidence to support this.

Sound Localization
One major task of the auditory system is to calculate the position from which a sound
is coming. This has important survival value for all animals and is calculated by the
brain from a comparison of the way the same sound appears to each of the two ears. It
is only necessary to switch the hi-fi between mono and stereo to appreciate how much
information we can gain from this. Two major mechanisms have been proposed for
sound localization; these are based on 1) a determination of the difference in the time
at which the same sound reaches each ear (the interaural time delay; fig. 9-14A) and
2) the difference in intensity between the sound in each ear (the interaural intensity
difference—fig. 9-14B).
Differences in the time at which a sound reaches each ear are extremely small and
their detection is a major challenge for the nervous system. If a sound comes from
 Hearing and the Processing of Musical Sound by the Brain 319

Figure 9-14. Sound localization. The brain has two ways of working out where a sound is com-
ing from. A. The first uses the difference in time between a sound reaching the left and the right
ears (interaural time delay). When the sound is coming from directly in front or behind, it reaches
both ears simultaneously. The maximum time difference (when the sound is coming from the
right or left) is 0.6 milliseconds, and lower delays code for the intervening angles. This method of
sound localization only works if the wavelength is greater than the width of the head. For shorter
wavelengths (higher frequencies) where distance between the two ears is equivalent to two or
more wavelengths, time delay does not give a unique solution for sound direction. Under these
circumstances, differences in intensity of the sound reaching the two ears is used (B). The sound
shadow created by the head is greatest for sources coming from the right or left and falls to zero
for sounds that originate from directly in front or behind.

directly in front or behind the head, it reaches both ears simultaneously, but if it comes
from the side, the maximum time difference is only 0.6 milliseconds (six ten-thou-
sandths of a second), which is the time sound takes to travel the 20 centimeters or so
that separate the ears. Of course in your dog, hamster, or budgerigar this distance is
much smaller and the task correspondingly more difficult! The start of a sound is a
clear signal whose arrival time can be easily compared. For a continuous sound the
situation is more complicated, and the interaural time delay must be measured from
320 Chapter Nine

the same point on the sound wave (e.g., the peak amplitude) in both ears. If the wave-
length is smaller than the distance between the ears, the signal will be ambiguous for
continuous sound because it will depend on how many cycles of the waveform lie
between the peaks detected in each ear. Sounds of 3,000 Hz and above (top G on the
piano) can fit two or more cycles in that distance and so the system that uses interaural
time delay starts to generate ambiguous results at about 1,500 Hz, which is an octave
below. For higher frequencies, differences in sound intensity must therefore be used
for sound source localization. Sounds coming from directly ahead or behind will again
be equally loud in both ears, but for those coming from the side, the head creates a
sound shadow so that the sound appears quieter in the ear on the opposite side from
the source. This mechanism does not work for low frequencies because the longer
wavelength sounds can bend around the head, reducing the perceived intensity differ-
ence. Interaural time delays and intensity differences do not allow any discrimination
of the elevation of the sound source above the horizon. This appears to be achieved
by analyzing the distortion of the sound by the external ear, which selectively alters
different frequency components depending on the vertical position of the source. Ste-
reophonic effects in the horizontal plane can readily be generated by recording sound
from two microphones placed at different positions on the stage, but this cannot be
used to produce the illusion of sound sources at different heights.
The accuracy of sound localization in the horizontal plane is quite remarkable, typi-
cally in the range of 3°–10°. This can be improved with practice and consequently
tends to be better in blind people (Roder et al. 1999). Interestingly, research has shown
that conductors have unusually good sound localization abilities. When their perfor-
mance in sound localization tasks was compared with musicians such as pianists and
nonmusicians, all groups performed equally well when the sound sources were more
or less directly ahead, but the conductors were much better at localizing sounds from
either side. It has been suggested that this superior performance may be developed
through learning to use spectral differences in sound coming from different directions.
This would involve detecting changes generated both by the filtering properties of the
external ear and by the head shadow (Munte et al. 2001). From brain scanning studies,
it has been possible to show that the cortical analysis of the spatial organization of
the auditory world is carried out in the posterior region of the planum temporale at a
position that is quite distinct from the regions that interpret the sequential changes in
pitch that underlie melody (Warren and Griffiths 2003). Spatial analysis of the sound
world also occurs outside the cortex in the inferior colliculi, which forms part of the
roof of the brain stem (fig. 9-11).

Brain Circuitry Underlying Emotional Responses to Music

Music activates not only the parts of the brain that overtly analyze its structure and
syntax, but as we all well know from personal experience, it also influences areas
that generate emotional responses. Indeed, the main reason we listen to music is for
its emotional impact. Of course, our emotional responses to a piece of music may
be due as much to the associations it has for us as to its musical structure. One only
has to think of the strong reactions that can be generated by the dire music of most
 Hearing and the Processing of Musical Sound by the Brain 321

national anthems to realize that it is not always the quality of the composition that is
the most important factor. To give another example, being Scottish, I find the sound
of the bagpipes both stirring and highly evocative and likely to produce feelings of
nostalgia, while the same sound once prompted a young relative of mine to complain
to her mother that she didn’t like “the men playing the Hoovers”! For most composers,
however, success depends on their ability to manipulate the emotions of the audience
directly through the music. This requires that particular musical elements produce pre-
dictable effects on the brain (Peretz, Gagnon, and Bouchard 1998). We may not all be
touched by the same pieces, but within the Western musical tradition at least, warlike
pieces tend to be strongly rhythmic and sometimes discordant, love songs harmonious
and legato, happy music lively and in major keys, and sad pieces slower and in minor
ones. Many other sensory stimuli also generate strong emotions, for example, the sight
of particular facial expressions, the intonation of the voice, a smell, or a taste; but it has
been argued that music differs from these in that it is an abstract phenomenon that does
not intrinsically carry signals from our environment that are important for survival
(Blood et al. 1999). While this is to some extent true, our responses to the other sen-
sory signals, such as smell, are often as much related to learned associations as to their
objective information content. While we may all be repelled by the odor of spoiled
food (a response with an obvious survival value), we may as individuals be comforted
by familiar odors specific to the environment in which we grew up, odors that to oth-
ers may appear neutral or even disquieting. In this regard, the response to music is not
so different; however, the personal nature of our reactions can make studying them
difficult. One attempt to do this compared the reaction of the brain to different degrees
of dissonance, which is a quality most listeners can agree upon as being unpleasant.
The research revealed consistent differences in the pattern of brain activity induced by
consonance and dissonance in regions thought to be involved in emotional responses
(Blood et al. 1999). However, music sometimes leads to brief positive responses
that are not just mildly pleasant, but positively euphoric. This can induce a physical
sensation that may be either visceral or feel like a “shiver down the spine.” Both of
these types of sensation imply the involvement of the autonomic nervous system (see
chapter 10), which engenders responses over which we have no conscious control. The
autonomic nervous system supplies the visceral organs and also the blood vessels and
glands, including structures in the skin such as sweat glands and the minute muscles
that cause the hairs to rise (piloerection). It is probably the action of the piloerector
muscles that is responsible for the aforementioned shivers. When music is able to
generate such reactions, they are accompanied by detectable increases in heartbeat
and breathing rate. This provides an objective verification of the timing and genuine-
ness of the euphoric feelings and so enables them to be correlated with brain activity.
The brain regions concerned are the same ones that are activated by other hedonistic
stimuli such as food, sex, and drugs of abuse (Blood and Zatorre 2001). They are part
of a reward system within the central nervous system that includes structures known
to be connected to the autonomic nervous system but is quite distinct from the regions
activated by mildly pleasant musical stimuli such as consonance.
In order to appreciate how music achieves its emotional impact on the brain, it is
first necessary to understand something of the organization of the circuitry that has
322 Chapter Nine

Figure 9-15. Some of the principal regions of the brain involved in emotional responses to music
and other stimuli are shown here. These lie mainly in the ventral regions of the brain and include
several elements of the limbic system, for example, the cingulate gyrus—particularly its anterior
part; the amygdala and hippocampus, which lie at the tip of the temporal lobe; and the ventral
prefrontal cortex. The ventral tegmental area of the midbrain contains nerve cells that send their
axons forward to the thalamus, hypothalamus, ventral striatum, and nucleus accumbens. Activity
in this latter structure is strongly linked to pleasurable emotions. Emotional responses also involve
the orbitofrontal cortex, which lies just in front of the nucleus accumbens. The arrows indicate
some of the major connections between these regions.

been implicated in pleasurable and unpleasurable sensations. Unfortunately, this is

a complex area, not least because it is still incompletely understood, and even the
simplified outline presented here may initially appear rather impenetrable. However,
as we prize music mainly for its emotional impact, some readers may wish to delve
further into this subject, and with a little perseverance this introduction should give
them the means to make some sense of the original literature. The parts of the brain
discussed below are illustrated in figure 9-15.
The pathways that mediate pleasure originate mainly in the ventral regions of the
brain. Some of the centers involved lie in the brain stem, particularly within the mid-
brain. One key element is a region known as the ventral tegmental area. This lies on
the midline of the midbrain and is populated by nerve cells that use dopamine as their
neurotransmitter. The axons of these neurons course forward toward the thalamus,
hypothalamus, and limbic system (see chapter 7) and are consequently described as
mesolimbic fibers (meso = mid). The parts of the limbic system most strongly linked
to emotional responses are the amygdala, which contributes the emotional coloring
to memory, and the anterior part of the cingulate gyrus, an area of the cortex that
 Hearing and the Processing of Musical Sound by the Brain 323

sits immediately above the corpus callosum. An important target of the mesolimbic
pathway is a structure known as the nucleus accumbens, which sits at the front of the
basal ganglion complex. Some studies refer to this general area as the ventral striatum,
because the corpus striatum is another name for the components of the basal ganglia
that are found here. The nucleus accumbens is linked to the area of the cortex that sits
immediately anterior (in front) of it. This is known as the orbitofrontal cortex, as it
lies immediately above the part of the skull that forms the roof of the eye socket (or
orbit). Though this is not an exhaustive list of the areas that are implicated in positive
and negative emotions, it is sufficient to support a general description of the effects
of music.
Some insight into the pleasure network of the brain in humans has been gathered
from subjective accounts of people who, for various clinical reasons, have had stimu-
lating electrodes implanted in a few of the areas listed in the previous paragraph. From
the accounts of these patients it is clear that we need to distinguish between wanting
(feeling compelled) to do something and liking (gaining pleasure from) it (Berridge
2003). Stimulation of the mesolimbic dopamine pathway is compulsive, and it has
been suggested that activity in these fibers is linked to addiction. Stimulation of the
nucleus accumbens, on the other hand, appears to evoke pleasurable sensations. This
is dependent on the release of opiatelike chemicals (e.g., enkephalins and endorphins)
that act as neurotransmitters (see chapter 1). Activity in the orbitofrontal lobes and in
the anterior cingulate gyrus is often associated with pleasure but may also be involved
in negative emotions. The physical manifestations of these emotions are largely
brought about by the hypothalamus and to some extent by a region of the cortex
known as the insula, which lies tucked out of sight behind the temporal lobe. Both of
these exert a controlling influence over the autonomic nervous system. Having gained
an overview of the effect of pleasurable sensations on the brain, we can now move on
to consider the emotional consequences of music.
If we return to the studies of Blood and Zatorre, we can now appreciate their obser-
vations that during music that induced the “chills” in their subjects, increased activity
was observed in the midbrain, the ventral striatum, the orbitofrontal cortex, and the
insula but activity in the amygdala was reduced. The imaging system used here had a
rather low spatial resolution, so it was not possible to specifically identify the nucleus
accumbens due to its small size. However, similar studies carried out at higher resolu-
tion not only confirmed this overall pattern of brain activity but clearly demonstrated
the involvement of the nucleus accumbens (Menon and Levitin 2005; Brown, Marti-
nez, and Parsons 2004). When the effects of “happy” and “sad” music were compared,
the former activated the ventral striatum and anterior cingulate gyrus, while the latter
produced activity in the hippocampus and amygdala (Mitterschiffthaler et al. 2007).
The introduction of dissonance into musical stimuli also led to increased activity in
the amygdala (Koelsch et al. 2006). This is consistent with its well-known role in the
recognition of negative emotion (particularly fear) in the facial expression of others.
Its response to the emotional content of music has been further explored through ob-
servations on a patient in whom it was completely destroyed on both sides of the brain,
while adjacent regions of the temporal cortex involved in music processing were pre-
served (Gosselin et al. 2007). In comparison to normal control subjects, the ability to
324 Chapter Nine

label musical extracts correctly as “scary” or “sad” was considerably impaired, while
the recognition of music as “happy” or “peaceful” was much more close to normal.
This was not due to a failure to extract relevant aspects of musical structure, as the
response to tempo and mode remained intact. The reward system evolved to motivate
animals to carry out behavior important for survival. Though this is not the case for
music, its ability to tap into this system in a nondetrimental way allows it to play a
significant role in supporting our emotional well-being.

The “Mozart Effect”

There has been much debate concerning whether music can improve mental perfor-
mance. In 1993, Rauscher et al. published a short letter in the prestigious scientific
journal Nature that purported to show that a brief period of listening to Mozart re-
sulted in improved performance in certain tasks forming one element of a battery of
standard intelligence tests. This resulted in a considerable degree of fevered press
speculation along the lines that listening to the music of Mozart or other socially
acceptable classical composers increases intelligence. It is only through the many at-
tempts by other researchers to replicate the effect that the truth has slowly emerged.
The claims of the original papers by Rauscher et al. were rather modest (Rauscher,
Shaw, and Ky 1993, 1995). A group of college students spent ten minutes listening
to a piece of music by Mozart (the first movement [Allegro] of the Sonata for Two
Pianos, K. 448) or to taped relaxation instructions, or spent the same length of time
sitting in silence. The precise nature of the movement of the Mozart sonata used turns
out to be crucial; it is of course fast but also in a major key, with a lively interaction
between the two instruments as they pass elements of the melody back and forth.
After listening to the music or sitting in silence, the subjects of the study were asked
to perform a spatial reasoning test called the Paper Folding and Cutting (PF&C) task.
This involves predicting the pattern produced if a piece of paper is folded repeatedly
and small sections are then cut out, as might be done to make a symmetrical Christmas
snowflake decoration. Interestingly, tasks of this nature should show a gender bias in
favor of men (Kimura 1999), but most studies of the Mozart effect do not mention
the sex composition of the experimental or control groups. The results obtained by
Rauscher suggested that the students who listened to Mozart performed slightly better
than the control groups on the PF&C task but that this was short-lived, and neither
these nor any subsequent studies have found any evidence of an enhancement in other
measures of intelligence or cognitive ability. The idea behind the original experiments
was based on some interesting but unproven theories suggesting that higher brain
function is dependent on how patterns of activity in small columns of nerve cells
within the cortex change over time. The rationale runs as follows: 1) because com-
puter models of these theoretical patterns (which might represent fundamental neural
activity) can be used to generate music-like sounds, responses to music early in life
may be based on them; 2) Mozart was composing at a very early age, so his music in
particular may express features of this patterning particularly strongly; 3) the PF&C
task also involves analyzing patterns that change with time and which might, by some
unspecified mechanism, be improved by unspecified intrinsic properties of Mozart’s
 Hearing and the Processing of Musical Sound by the Brain 325

music that reflect the relevant mental activity. Needless to say, this is a rather tenuous
line of reasoning. Apart from the flaws in the underlying logic, there are problems
with the experimental design, as many subsequent researchers who have attempted
to replicate the experiments have pointed out. The effect of the music was compared
to the effect of silence, which is unstimulating by comparison. Music is an extremely
complex stimulus; even pieces of music by the same composer will generate very dif-
ferent emotions depending on their structure. It is vital that such factors are controlled
for in experiments that investigate this question. Mozart appears to have been chosen
for the reputation he has with a (lamentably perhaps!) small proportion of the general
population who mostly share a common cultural background. The fact that he was
twenty-five at the time he composed K. 448 in 1781 and that his style had developed
very much in keeping with prevailing cultural trends is not recognized by logic back-
ing the theory. Other composers or (heaven forbid) other musical styles were not
compared in the original study. It is hard to imagine school boards pushing the music
of Led Zeppelin had this been shown to improve performance in the PF&C test! We
know nothing of the cultural background or musical tastes of the control group in the
original study, which could have significantly influenced their response to the music
and hence perhaps their performance in the test. Many studies that have tried to repli-
cate the Mozart effect have failed (Chabris 1999; Steele et al. 1999), though it must be
recognized that a significant proportion of them have severe methodological flaws of
their own. In several, however, attempts were made to address some of the problems
of the original experimental design, for example, by comparing the effect of music
by Mozart with that of other classical or nonclassical composers or with listening to a
story. The results of some studies suggested that improvement in the PF&C task was
related to listening to a preferred sensory experience, whether musical or not (Nantais
and Schellenberg 1999) and was related to the mood or degree of arousal that this
induced (Thompson, Schellenberg, and Husain 2001; Nantais and Schellenberg 1999).
In a telling experiment, a comparison was made between the effects of K. 448 played
at the original tempo or at a much slower one, and of a version in a minor key played
at a fast or slow tempo. The same conclusion was reached (Husain, Thompson, and
Schellenberg 2002).
Given the often frenzied attention that has been (and is still being) given to the “Mo-
zart effect” and its continued promotion in the popular media, it is only fair to give
the last word to the only author of the original paper who continued to study it: “The
Mozart effect was studied only in adults, lasted only a few minutes, and was found
only for spatial-temporal reasoning. Nevertheless the effect has spawned a Mozart
industry . . . claiming that listening to classical music can make children ‘smarter.’ In
fact no scientific evidence supports (this) claim. . . . Although the Mozart effect is of
scientific interest, its education implications appear to be limited” (Rauscher 2003).
The broader question of whether acquiring practical musical skills, as opposed to
passive listening, exerts any educational influence on nonmusical abilities in children
is an equally controversial issue. Isolating the effect of musical training, per se, from
the stimulation of being in a select group receiving special attention, or the effect of
coming from a background that supports involvement in such schemes, is difficult
to achieve in an ethical way. A good review of research in this area can be found in
326 Chapter Nine

Schellenberg (2006). The best that can be said, once these extraneous factors have
been eliminated, is that the benefits appear marginal and may not be long-lasting.
Given that an active participation in any form of music making is such a personally
enriching experience, the need to justify it on other grounds seems rather to miss the
point, particularly if this is linked to a narrow cultural agenda. What is clear is that
providing good support for music, of whatever genre, within the educational system,
will lead to more practicing musicians within the community, and this should be a
worthy objective both for the individual and for society.

Epilepsy
The effect of Mozart’s music on brain activity linked to epilepsy has also received
some attention. This is rather more amenable to objective study because epilepsy is
associated with clearly identifiable patterns of synchronized electrical activity in the
brain that can easily be recorded and analyzed statistically. There is evidence from
one group that listening to Mozart’s K. 448 can reduce the frequency of the electrical
activity associated with epileptic seizures (Turner 2004; Hughes 2001, 2002). This
piece appears to be more effective than several other pieces of music, including those
by other classical composers such as Beethoven; however, the number of patients who
have been studied and the number of pieces of music that have been tested are both
very small, so much more work is needed to validate the effect. Hughes analyzed the
structural features of the music of several composers and, based on the characteris-
tics of K. 448, has made certain predictions about which pieces by Mozart and other
composers might be most useful in controlling seizures. For example, he considered
that the music of Haydn would be a good candidate, though it might not be quite as
effective as that of Mozart. Unfortunately, his hypothesis remains as yet untested.
Interestingly, music can also induce epileptic attacks in some individuals (Kaplan
2003). The nineteenth-century Chinese poet Kung Tzu-chen gives what appears to
be an early account of music-induced epilepsy: “Since my remote boyhood I have
always been absentminded while hearing the sound of the street vendor’s flute. I fall
sick when I hear the sound of the flute in the evening sun, although I do not know
the reason.” The suggestion by some authors that the following Shakespearean quote
from The Merchant of Venice (act IV, scene 1) also refers to musicogenic epilepsy is
perhaps on shakier ground (Wieser et al. 1997):

Some men there are, love not a gaping pig;

Some are mad, if they behold a cat;
And others, when the bagpipe sings i’ the nose
Cannot contain their urine.

Though the ability of flashing lights of certain frequencies is well known to trigger
epileptic seizures, there is no evidence that rhythm or any other elements of musical
structure consistently have this effect. Instead, the type of music (if any) likely to in-
duce an attack is highly specific to the individual. In most cases, the musical stimulus
is of external origin but in some patients an attack can be induced by their own sing-
 Hearing and the Processing of Musical Sound by the Brain 327

ing. Music of all styles (classical orchestral or opera, religious, folk, jazz, military,
dance, music evoking a particular mood or of cultural significance, etc.) has been
known to trigger attacks, but the underlying mechanisms remain unknown. Pure tones
do not induce seizures, and the crucial aspects of musical stimuli (which may include
its emotional impact) suggest that the attack does not arise from activity in the primary
auditory cortex, but in regions where higher-order processing of music takes place.

REFERENCES

Altenmuller, E. O. 2001. How many music centers are in the brain? Ann NY Acad Sci 930:273–
80.
Audsley, G. A. 1965. The art of organ-building: A comprehensive historical, theoretical, and
practical treatise on the tonal appointment and mechanical construction of concert-room,
church, and chamber organs. New York: Dover.
Ayotte, J., I. Peretz, and K. Hyde. 2002. Congenital amusia: A group study of adults afflicted
with a music-specific disorder. Brain 125 (Pt 2):238–51.
Bangert, M., T. Peschel, G. Schlaug, M. Rotte, D. Drescher, H. Hinrichs, H. J. Heinze, and
E. Altenmuller. 2006. Shared networks for auditory and motor processing in professional
pianists: Evidence from fMRI conjunction. Neuroimage 30 (3):917–26.
Behar, A., W. Wong, and H. Kunov. 2006. Risk of hearing loss in orchestra musicians: Review
of the literature. Med Prob Perform Artists 21:164–68.
Benade, A. H. 1973. The physics of the brasses. Sci Am 229 (July):24–35.
Bendor, D., and X. Wang. 2005. The neuronal representation of pitch in primate auditory cor-
tex. Nature 436 (7054):1161–65.
———. 2006. Cortical representations of pitch in monkeys and humans. Curr Opin Neurobiol
16 (4):391–99.
Berridge, K. C. 2003. Pleasures of the brain. Brain Cogn 52 (1):106–28.
Bicknell, S. 1999. Harmonics and “Cheats,” at www.stephenbicknell.org/3.6.01.php (accessed
April 23, 2008).
Blood, A. J., and R. J. Zatorre. 2001. Intensely pleasurable responses to music correlate with
activity in brain regions implicated in reward and emotion. Proc Natl Acad Sci USA 98
(20):11818–23.
Blood, A. J., R. J. Zatorre, P. Bermudez, and A. C. Evans. 1999. Emotional responses to pleas-
ant and unpleasant music correlate with activity in paralimbic brain regions. Nat Neurosci
2 (4):382–87.
Brown, M. C. 1999. Audition. In Fundamental neuroscience, ed. M. J. Zigmond, F. E. Bloom,
S. C. Landis, J. L. Roberts, and L. R. Squire. San Diego: Academic Press.
Brown, S., and M. J. Martinez. 2007. Activation of premotor vocal areas during musical dis-
crimination. Brain Cogn 63 (1):59–69.
Brown, S., M. J. Martinez, D. A. Hodges, P. T. Fox, and L. M. Parsons. 2004. The song system
of the human brain. Brain Res Cogn Brain Res 20 (3):363–75.
Brown, S., M. J. Martinez, and L. M. Parsons. 2004. Passive music listening spontaneously
engages limbic and paralimbic systems. Neuroreport 15 (13):2033–37.
Burns, E. M. 1999. Intervals, scales and tuning. In The psychology of music, ed. D. Deutsch.
San Diego: Academic Press.
Chabris, C. F. 1999. Prelude or requiem for the “Mozart effect”? Nature 400 (6747):826–27;
author reply 827–28.
328 Chapter Nine

Chen, J. L., V. B. Penhune, and R. J. Zatorre. 2008. Moving on time: Brain network for audi-
tory-motor synchronization is modulated by rhythm complexity and musical training. J Cogn
Neurosci 20 (2):226–39.
Dronkers, N. F., O. Plaisant, M. T. Iba-Zizen, and E. A. Cabanis. 2007. Paul Broca’s historic
cases: High resolution MR imaging of the brains of Leborgne and Lelong. Brain 130 (Pt
5):1432–41.
Evers, S., J. Dannert, D. Rodding, G. Rotter, and E. B. Ringelstein. 1999. The cerebral hae-
modynamics of music perception. A transcranial Doppler sonography study. Brain 122 (Pt
1):75–85.
Fletcher, N. H., and T. D. Rossing. 1998. The physics of musical instruments. 2nd ed. New
York: Springer.
Foxton, J. M., J. L. Dean, R. Gee, I. Peretz, and T. D. Griffiths. 2004. Characterization of defi-
cits in pitch perception underlying “tone deafness.” Brain 127 (Pt 4):801–10.
Fujioka, T., B. Ross, H. Okamoto, Y. Takeshima, R. Kakigi, and C. Pantev. 2003. Tonotopic
representation of missing fundamental complex sounds in the human auditory cortex. Eur J
Neurosci 18 (2):432–40.
Gaser, C., and G. Schlaug. 2003. Brain structures differ between musicians and non-musicians.
J Neurosci 23 (27):9240–45.
Gazzola, V., L. Aziz-Zadeh, and C. Keysers. 2006. Empathy and the somatotopic auditory mir-
ror system in humans. Curr Biol 16 (18):1824–29.
Gosselin, N., I. Peretz, E. Johnsen, and R. Adolphs. 2007. Amygdala damage impairs emotion
recognition from music. Neuropsychologia 45 (2):236–44.
Grahn, J. A., and M. Brett. 2007. Rhythm and beat perception in motor areas of the brain. J
Cogn Neurosci 19 (5):893–906.
Griffiths, T. D. 2001. The neural processing of complex sounds. Ann NY Acad Sci 930:133–42.
Griffiths, T. D., and J. D. Warren. 2002. The planum temporale as a computational hub. Trends
Neurosci 25 (7):348–53.
Halpern, A. R., R. J. Zatorre, M. Bouffard, and J. A. Johnson. 2004. Behavioral and neural cor-
relates of perceived and imagined musical timbre. Neuropsychologia 42 (9):1281–92.
Hart, C. W., M. A. Gletman, J. Schubach, and M. Santucci. 1987. The musician and occupa-
tional sound hazards. Med Prob Perform Artists 2:22–25.
Henderson, D., E. C. Bielefeld, K. C. Harris, and B. H. Hu. 2006. The role of oxidative stress
in noise-induced hearing loss. Ear Hear 27 (1):1–19.
Henoch, M. A., and K. S. Chesky. 1999. Ear canal resonance as a risk factor in music-induced
hearing loss. Med Prob Perform Artists 14:103–6.
Hughes, J. R. 2001. The Mozart effect. J R Soc Med 94 (6):316.
———. 2002. The Mozart effect: Additional data. Epilepsy Behav 3 (2):182–84.
Husain, G., W. F. Thompson, and E. G. Schellenberg. 2002. Effects of music tempo and mode
on arousal, mood and spatial abilities. Music Perception 20 (2):241–58.
Hyde, K. L., J. P. Lerch, R. J. Zatorre, T. D. Griffiths, A. C. Evans, and I. Peretz. 2007.
Cortical thickness in congenital amusia: When less is better than more. J Neurosci 27
(47):13028–32.
Hyde, K. L., R. J. Zatorre, T. D. Griffiths, J. P. Lerch, and I. Peretz. 2006. Morphometry of the
amusic brain: A two-site study. Brain 129 (Pt 10):2562–70.
Jamison, H. L., K. E. Watkins, D. V. Bishop, and P. M. Matthews. 2006. Hemispheric special-
ization for processing auditory nonspeech stimuli. Cereb Cortex 16 (9):1266–75.
Jaramillo, F., V. S. Markin, and A. J. Hudspeth. 1993. Auditory illusions and the single hair
cell. Nature 364 (6437):527–29.
 Hearing and the Processing of Musical Sound by the Brain 329

Kaas, J. H., T. A. Hackett, and M. J. Tramo. 1999. Auditory processing in primate cerebral
cortex. Curr Opin Neurobiol 9 (2):164–70.
Kaplan, P. W. 2003. Musicogenic epilepsy and epileptic music: A seizure’s song. Epilepsy
Behav 4 (5):464–73.
Keenan, J. P., V. Thangaraj, A. R. Halpern, and G. Schlaug. 2001. Absolute pitch and planum
temporale. Neuroimage 14 (6):1402–8.
Kim, K. H., N. R. Relkin, K. M. Lee, and J. Hirsch. 1997. Distinct cortical areas associated with
native and second languages. Nature 388 (6638):171–74.
Kimura, D. 1999. Sex and cognition. Cambridge, Mass.: MIT Press.
Koelsch, S. 2006. Significance of Broca’s area and ventral premotor cortex for music-syntactic
prcessing. Cortex 42:518–20.
Koelsch, S., T. Fritz, D. Y. v. Cramon, K. Muller, and A. D. Friederici. 2006. Investigating
emotion with music: An fMRI study. Hum Brain Mapp 27 (3):239–50.
Kohler, E., C. Keysers, M. A. Umilta, L. Fogassi, V. Gallese, and G. Rizzolatti. 2002. Hear-
ing sounds, understanding actions: Action representation in mirror neurons. Science 297
(5582):846–48.
Kubba, A. K., and M. Young. 1996. Ludwig van Beethoven: A medical biography. Lancet 347
(8995):167–70.
Levitin, D. J., and S. E. Rogers. 2005. Absolute pitch: Perception, coding, and controversies.
Trends Cogn Sci 9 (1):26–33.
Liegeois-Chauvel, C., K. Giraud, J. M. Badier, P. Marquis, and P. Chauvel. 2001. Intracerebral
evoked potentials in pitch perception reveal a functional asymmetry of the human auditory
cortex. Ann NY Acad Sci 930:117–32.
Liegeois-Chauvel, C., I. Peretz, M. Babai, V. Laguitton, and P. Chauvel. 1998. Contribu-
tion of different cortical areas in the temporal lobes to music processing. Brain 121 (Pt
10):1853–67.
Limb, C. J., S. Kemeny, E. B. Ortigoza, S. Rouhani, and A. R. Braun. 2006. Left hemispheric
lateralization of brain activity during passive rhythm perception in musicians. Anat Rec A
Discov Mol Cell Evol Biol 288 (4):382–89.
Luders, E., C. Gaser, L. Jancke, and G. Schlaug. 2004. A voxel-based approach to gray matter
asymmetries. Neuroimage 22 (2):656–64.
Maess, B., S. Koelsch, T. C. Gunter, and A. D. Friederici. 2001. Musical syntax is processed
in Broca’s area: An MEG study. Nat Neurosci 4 (5):540–45.
Marin, O. S. M., and D. W. Perry. 1999. Neurological aspects of music perception and perfor-
mance. In The psychology of music, ed. D. Deutsch. San Diego: Academic Press.
Menon, V., and D. J. Levitin. 2005. The rewards of music listening: Response and physiologi-
cal connectivity of the mesolimbic system. Neuroimage 28 (1):175–84.
Menon, V., D. J. Levitin, B. K. Smith, A. Lembke, B. D. Krasnow, D. Glazer, G. H. Glover,
and S. McAdams. 2002. Neural correlates of timbre change in harmonic sounds. Neuroim-
age 17 (4):1742–54.
Mitterschiffthaler, M. T., C. H. Fu, J. A. Dalton, C. M. Andrew, and S. C. Williams. 2007.
A functional MRI study of happy and sad affective states induced by classical music. Hum
Brain Mapp 28 (11):1150–62.
Munte, T. F. 2002. Brains out of tune. Nature 415 (6872):589–90.
Munte, T. F., C. Kohlmetz, W. Nager, and E. Altenmuller. 2001. Neuroperception. Superior
auditory spatial tuning in conductors. Nature 409 (6820):580.
Nan, Y., T. R. Knosche, S. Zysset, and A. D. Friederici. 2008. Cross-cultural music phrase
processing: An fMRI study. Hum Brain Mapp 29 (3):312–28.
330 Chapter Nine

Nantais, K. M., and E. G. Schellenberg. 1999. The Mozart effect: An artifact of preference.
Psychol Sci 10:370–73.
Ohnishi, T., H. Matsuda, T. Asada, M. Aruga, M. Hirakata, M. Nishikawa, A. Katoh, and
E. Imabayashi. 2001. Functional anatomy of musical perception in musicians. Cereb Cortex
11 (8):754–60.
Palin, S. L. 1994. Does classical music damage the hearing of musicians? A review of the
literature. Occup Med (Lond) 44 (3):130–36.
Pantev, C., R. Oostenveld, A. Engelien, B. Ross, L. E. Roberts, and M. Hoke. 1998. Increased
auditory cortical representation in musicians. Nature 392 (6678):811–14.
Pantev, C., B. Ross, T. Fujioka, L. J. Trainor, M. Schulte, and M. Schulz. 2003. Music and
learning-induced cortical plasticity. Ann NY Acad Sci 999:438–50.
Parsons, L. M. 2001. Exploring the functional neuroanatomy of music performance, percep-
tion, and comprehension. Ann NY Acad Sci 930:211–31.
Passynkova, N., K. Sander, and H. Scheich. 2005. Left auditory cortex specialization for verti-
cal harmonic structure of chords. Ann NY Acad Sci 1060:454–56.
Patterson, R. D., S. Uppenkamp, I. S. Johnsrude, and T. D. Griffiths. 2002. The processing of
temporal pitch and melody information in auditory cortex. Neuron 36 (4):767–76.
Penagos, H., J. R. Melcher, and A. J. Oxenham. 2004. A neural representation of pitch salience
in nonprimary human auditory cortex revealed with functional magnetic resonance imaging.
J Neurosci 24 (30):6810–15.
Peretz, I. 2002. Brain specialization for music. Neuroscientist 8 (4):372–80.
Peretz, I., L. Gagnon, and B. Bouchard. 1998. Music and emotion: Perceptual determinants,
immediacy, and isolation after brain damage. Cognition 68 (2):111–41.
Peretz, I., and K. L. Hyde. 2003. What is specific to music processing? Insights from congenital
amusia. Trends Cogn Sci 7 (8):362–67.
Peretz, I., and R. J. Zatorre. 2005. Brain organization for music processing. Annu Rev Psychol
56:89–114.
Pierce, J. R. 1999. Pitch. In The psychology of music, ed. D. Deutsch. San Diego: Academic
Press.
Planchart, A. E. 1960. A study of the theories of Giuseppe Tartini. J Music Theory 4:32–61.
Pressnitzer, D., R. D. Patterson, and K. Krumbholz. 2001. The lower limit of melodic pitch. J
Acoust Soc Am 109 (5 Pt 1):2074–84.
Ramachandran, V. S., and E. M. Hubbard. 2003. Hearing colors, tasting shapes. Sci Am 288
(5):52–59.
Rasch, R., and R. Plomp. 1999. The perception of musical tones. In The psychology of music,
ed. D. Deutsch. San Diego: Academic Press.
Rauscher, F. H. 2003. Can music instruction affect children’s cognitive development? ERIC
Digest (September).
Rauscher, F. H., G. L. Shaw, and K. N. Ky. 1993. Music and spatial task performance. Nature
365 (6447):611.
———. 1995. Listening to Mozart enhances spatial-temporal reasoning: Towards a neurophysi-
ological basis. Neurosci Lett 185 (1):44–47.
Roder, B., W. Teder-Salejarvi, A. Sterr, F. Rosler, S. A. Hillyard, and H. J. Neville. 1999.
Improved auditory spatial tuning in blind humans. Nature 400 (6740):162–66.
Sacks, O. W. 2007. Musicophilia: Tales of music and the brain. New York: Alfred A. Knopf.
Sataloff, R. T. 1991. Hearing loss in musicians. Am J Otol 12 (2):122–27.
Schellenberg, E. G. 2006. Exposure to music: The truth about the consequences. In The child
as musician: A handbook of musical development, ed. G. E. McPherson. Oxford: Oxford
University Press.
 Hearing and the Processing of Musical Sound by the Brain 331

Schlaug, G., L. Jancke, Y. Huang, and H. Steinmetz. 1995. In vivo evidence of structural brain
asymmetry in musicians. Science 267 (5198):699–701.
Schneider, P., M. Scherg, H. G. Dosch, H. J. Specht, A. Gutschalk, and A. Rupp. 2002. Mor-
phology of Heschl’s gyrus reflects enhanced activation in the auditory cortex of musicians.
Nat Neurosci 5 (7):688–94.
Schreiner, C. E., and J. A. Winer. 2007. Auditory cortex mapmaking: Principles, projections,
and plasticity. Neuron 56 (2):356–65.
Shuter-Dyson, R. 1999. Musical ability. In The psychology of music, ed. D. Deutsch. San Di-
ego: Academic Press.
Steele, K. M., S. dalla Bella, I. Peretz, T. Dunlop, L. A. Dawe, G. K. Humphrey, R. A. Shan-
non, J. L. Kirby, Jr., and C. G. Olmstead. 1999. Prelude or requiem for the “Mozart effect”?
Nature 400 (6747):827–28.
Suteerawattananon, M., G. S. Morris, B. R. Etnyre, J. Jankovic, and E. J. Protas. 2004. Effects
of visual and auditory cues on gait in individuals with Parkinson’s disease. J Neurol Sci 219
(1–2):63–69.
Talavage, T. M., M. I. Sereno, J. R. Melcher, P. J. Ledden, B. R. Rosen, and A. M. Dale. 2004.
Tonotopic organization in human auditory cortex revealed by progressions of frequency
sensitivity. J Neurophysiol 91 (3):1282–96.
Thompson, W. F., E. G. Schellenberg, and G. Husain. 2001. Arousal, mood, and the Mozart
effect. Psychol Sci 12 (3):248–51.
Tramo, M. J., P. A. Cariani, B. Delgutte, and L. D. Braida. 2001. Neurobiological foundations
for the theory of harmony in western tonal music. Ann NY Acad Sci 930:92–116.
Turner, R. P. 2004. The acute effect of music on interictal epileptiform discharges. Epilepsy
Behav 5 (5):662–68.
Warr, W. B., and J. E. Boche. 2003. Diversity of axonal ramifications belonging to single lat-
eral and medial olivocochlear neurons. Exp Brain Res 153 (4):499–513.
Warren, J. D., and T. D. Griffiths. 2003. Distinct mechanisms for processing spatial sequences
and pitch sequences in the human auditory brain. J Neurosci 23 (13):5799–5804.
Watanabe, T., S. Yagishita, and H. Kikyo. 2008. Memory of music: Roles of right hippocam-
pus and left inferior frontal gyrus. Neuroimage 39 (1):483–91.
Wieser, H. G., H. Hungerbuhler, A. M. Siegel, and A. Buck. 1997. Musicogenic epilepsy:
Review of the literature and case report with ictal single photon emission computed tomog-
raphy. Epilepsia 38 (2):200–207.
Wright Reid, A. 1999. A sound ear, at www.symphony.dk/3info/sound-ear.htm (accessed April
23, 2008).
Yost, William A. 2000. Fundamentals of hearing: An introduction. 4th ed. San Diego: Aca-
demic Press.
Zatorre, R. J. 2001. Neural specializations for tonal processing. Ann NY Acad Sci 930:193–
210.
———. 2003. Absolute pitch: A model for understanding the influence of genes and develop-
ment on neural and cognitive function. Nat Neurosci 6 (7):692–95.
Zatorre, R. J., J. L. Chen, and V. B. Penhune. 2007. When the brain plays music: Auditory-
motor interactions in music perception and production. Nat Rev Neurosci 8 (7):547–58.
 Chapter Ten

Performance-Related Stress and

Its Management

The stress response is a natural reaction of the body that has evolved to prepare us for
extreme physical activity, usually in the face of a major threat from the environment or
from the creatures within it. In both animals and humans, this increases the chance of
escaping from the danger and as such is vital for survival. In our everyday lives, how-
ever, it may be triggered repeatedly by situations that are far from life threatening, and
this can interfere with our ability to function normally. This not only has emotional
consequences but reduces our control over activities that require precise motor skills.
Nowhere is this more evident than in performance-related stress (such as stage fright),
which is common among musicians, a group in which it has been studied extensively
(for a review see Smith, Maragos, and van Dyke 2000). Performance-related stress
can be broken down into two components. The first is known as state anxiety, mean-
ing that it is triggered by particular types of situations. However, the degree to which
an individual is susceptible to these situations will depend on his or her psychological
makeup. Those who are most vulnerable to it are said to exhibit trait anxiety, that is,
they have a predisposition to find the experience of performance threatening rather
than challenging or stimulating. Trait anxiety is most common in people who are self-
absorbed and have low self-esteem (Kendrick et al. 1982).
Stress can be acute, that is to say, a short-term response to isolated events. However,
it can also be a chronic state in which the body is maintained for long periods (days,
weeks, or months), and this can have serious consequences for general health. For
many people, the origin of stress is in the workplace, where they spend a major propor-
tion of their waking hours. Being at the mercy of an unreasonable or tyrannical boss or
conductor will produce stress in most people. Ironically, the behavior of the tyrant may
itself stem from a feeling of insecurity at being in the seat of command or from the wor-
ries about fulfilling the expectations or demands of superiors. Chronic stress may also
be a self-generated symptom of trait anxiety, arising from feelings of inadequacy. The
sufferer may feel that his or her performance is not up to standard or that colleagues are
critical of his or her abilities, regardless of whether this has any basis in fact.
Though stress is primarily associated in our minds with unpleasant situations, this
is by no means always the case. The thrill of anticipation before a first date is brought

332
 Performance-Related Stress and Its Management 333

Figure 10-1. This graph gives a schematic representation of the relation-

ship between stress and performance. Low levels of stress increase alertness
and prime the body for action, all of which can improve the quality of the
performance. As stress levels rise, however, the performer may experience a
loss of fine muscle control and experience muscle tremor. A number of other
symptoms such as dry mouth, rapid breathing, feelings of nausea, and failure
of short-term memory may appear, all of which conspire to reduce the ability
to perform effectively.

about by the physical manifestations of the stress response, while the attraction of
many rides at fairgrounds or theme parks or of extreme sports comes from the adrena-
line rush generated by brief exposure to a more or less controlled stressful situation.
Performance, whether musical or sporting, is likely to benefit from a degree of
stress. As we shall see, the stress response prepares the metabolic systems of the
body for activity and so is necessary for peak physical performance. If you go on-
stage feeling too relaxed or lethargic, your performance will remain lackluster, even
if it is technically very proficient. For a really excellent performance, not only must
all the resources for the physical aspects of playing or singing be at your command,
but it is also necessary to be mentally alert or charged. We can therefore imagine a
notional relationship between the level of stress and performance quality such as the
one shown in figure 10-1. Here quality increases with stress up to a certain level,
beyond which it declines as the physical symptoms of stress become increasingly
deleterious (Steptoe and Fidler 1987; Everly and Rosenfeld 1981). In this diagram,
the section labeled “stress” reflects not simply the physical changes that have been
wrought on the body, but how we as individuals respond to this psychologically. As
we become more experienced with performance or as we acquire strategies to deal
with it, the physical symptoms of performance-related stress may or may not decline,
334 Chapter Ten

but how we respond to or harness these changes to our advantage certainly does. This
chapter will therefore pay particular attention to a range of methods that can be used
to manage stress.

THE PHYSICAL SYMPTOMS OF STRESS

We are all familiar with the physical manifestations of acute stress; we experience
them during an audition, a job interview, or when we find ourselves in an emotional
or physical confrontation or a frightening situation. Our heart rate increases and may
become more forceful so that we feel it thumping in our chest. We start to sweat, but
because the skin has not yet become heated by the activity of the muscles beneath,
this sweat feels cold. We may feel the hair on the scalp or the rest of the skin ris-
ing, have the sensation of butterflies in the stomach, and even become nauseous.
Breathing becomes rapid and shallow and the mouth dry, and the muscles (includ-
ing those involved in breathing and controlling the hands) may tremble, making
the physical aspects of singing and playing difficult. In addition, the mind may go
blank (a failure of working memory) so that music previously safely committed to
memory becomes difficult or impossible to recall. If you have a vivid imagination,
you may already be feeling uncomfortable just from reading this list! With chronic
stress, additional symptoms may arise. One of these is called somatization, which is
the appearance of physical symptoms that, while perfectly real to the sufferer, may
have no organic basis. These symptoms can include pain (e.g., in the arm, chest, or
abdomen), headaches, or fatigue. Somatization is one reason why accurate diagnosis
of musicians’ ailments (e.g., hand or arm pain) can be difficult and needs to be based
on a complete case history and an appreciation of all of the relevant lifestyle factors.
Without this, there is a risk that inappropriate and potentially damaging treatment
may be advocated. Chronic stress may also lead to a reduction in the effectiveness
of the immune system, resulting in an increased susceptibility to infection (Lovallo
2005). For example, studies of adolescents and medical students under exam and
other stressful conditions have shown that they have an increased chance of falling
victim to viral infections such as herpes, glandular fever (mononucleosis), or throat
infections.
Unrelieved stress can in the long term induce or aggravate a number of other health
problems. The increased demands on the heart and the constriction of blood vessels
to the organs can contribute to high blood pressure (hypertension), the occurrence of
migraines, and heart disease. Stress can have a number of effects on the gastrointes-
tinal system. It can lead to excessive acid secretion in the stomach, which may con-
tribute to the development of gastric ulcers, and aggravate irritable bowel syndrome
and other related diseases. It can be associated with gastroesophageal reflux, in which
the acid contents of the stomach rise into the esophagus. This causes heartburn and
irritates the lining of the esophagus and may have particular consequences for singers
(see chapter 5). It is not uncommon for asthma and skin conditions such as eczema to
flare up. Finally, as discussed below, chronic stress can also trigger general or clinical
depression.
 Performance-Related Stress and Its Management 335

HOW THE NERVOUS SYSTEM GENERATES

THE STRESS RESPONSE

The reaction of the nervous system to stress can be divided into two phases. The first
is responsible for the recognition that a situation is threatening or stressful, while the
second produces the appropriate physiological changes needed to cope with this. We
will address these in turn.

The Limbic System

The recognition of a situation as being potentially stressful is the responsibility of the
limbic system. This is a network composed of a number of areas in the forebrain (fig. 7-
4C). We frequently worry about certain forthcoming events because of past experiences,
so it should be no surprise to discover that two areas of the cortex that are involved in
memory belong to the limbic system. One (the hippocampus) is responsible for laying
down what is known as declarative memory—the memory of facts, events, and places,
while another region (the amygdala) is responsible for the emotional aspects of memory.
Activity in the amygdala is particularly associated with negative emotions such as fear.
Both types of memory are inextricably linked to our interpretation of future situations.
Clearly the emotional content of a memory will play a significant role in determining
whether we regard situations as posing a threat or not. For example, hearing a particular
piece of music may have little or no significance for one person, while for another, it
may trigger strong positive (e.g., romantic) or negative (apprehensive) emotions depend-
ing on the circumstances in which it has been previously encountered. The hippocampus
makes the association, while the amygdala indicates the emotional coloring. Associated
with the limbic system are regions of the prefrontal cortex (the part of the brain lying
just behind the forehead) where we plan our future actions. Through our imaginations
we can construct a form of virtual reality through which we can play out various pos-
sible scenarios in order to consider what their consequences might be. This makes use
of emotional memories and so, just by thinking about how a forthcoming performance
may go, we can generate real physical responses. What we imagine will, of course, be
shaped by past experience. If we have had a previous performance failure, we will worry
more, whereas if our experiences were good, we will be more relaxed. These imagined
scenarios may be quite irrational. For example, our apprehension may be greater at a
venue where we once had a bad experience, even though everything has gone well since,
but this doesn’t make the feelings any less real. Clearly, the danger is that overanxiety
about a negative outcome may contribute to the prophecy becoming self-fulfilling. This
negative outlook is sometimes known as catastrophization—imagining only the worst
possible outcome. On the other hand, with help it should also be possible to use our
mental resources to reduce stress by reinterpreting potentially threatening situations that
we know we will have to encounter.

The Hypothalamus and the Autonomic Nervous System

The physical manifestation of the emotional responses generated by the limbic system
is controlled mainly by the hypothalamus (fig. 7-4A). This is the major center in the
336 Chapter Ten

brain for maintaining what is known as homeostasis; that is, it regulates the metabolic
processes of the body so that they are set at a level that is appropriate for the current
situation. This control is exerted through the action of nerves and hormones. If the
body becomes cold, mechanisms are set in motion by the hypothalamus to increase
its temperature by mobilizing energy reserves and increasing muscular activity (in-
cluding shivering) and through a reduction of heat loss by diverting blood away from
the skin. It may also stimulate appetite to increase food intake. If the body becomes
dehydrated, the hypothalamus increases fluid retention by the kidneys and generates
a feeling of thirst to encourage drinking. In terms of the stress response, the crucial
factor is that the hypothalamus also has the ability to switch body metabolism from a
state that is compatible with passivity to one that will support the vigorous activity ap-
propriate when facing a threat. It achieves its objectives through two executive arms.
The first is the autonomic nervous system, which is described below. The second is
through the release of hormones into the bloodstream, either from the pituitary gland,
which hangs from the underside of the brain just below the hypothalamus, or from
other hormone-secreting glands within the body. For the stress response, the most im-
portant hormonal response comes from what is known as the hypothalamic pituitary
adrenal cortical system. Under the control of the hypothalamus, the pituitary gland
releases a signal molecule into the bloodstream that activates the adrenal glands (fig.
10-2). These sit like a cap on top of the kidneys and are composed of two parts. The
cores of the glands secrete adrenaline (which we will consider later), while the outer
part (the cortex) secretes into the bloodstream hormones such as cortisol that play a
major role in the stress response. Cortisol promotes the formation of glucose, which is
the fuel that cells burn to produce energy from reserves stored in the body. However,
it also suppresses the immune system, and though this is of little significance in the
short term, we have seen that it may cause problems in chronic stress.
The autonomic nervous system regulates the distribution of blood to the muscles,
organs, and skin and also controls the activity of glands such as those producing
saliva, sweat, and digestive enzymes. It has two main divisions, known as the sympa-
thetic and the parasympathetic systems (fig. 10-2). The parasympathetic system exerts
its actions through many of the cranial nerves (see chapter 7) and through nerves that
emerge from the lower (sacral) part of the spinal cord. When we are slumped on the
sofa in front of the fire after a heavy meal, it is the parasympathetic nervous system
that is ascendant. Under these circumstances, blood flow is directed to the gut, and
we feel drowsy because less is going to the brain. Digestion is promoted through the
secretion of saliva into the mouth and digestive enzymes in the gut, while the muscular
walls of the stomach and intestines contract rhythmically to mix the food and enzymes
together. The heart rate slows, the pupils become smaller, and the bronchioles in the
lungs narrow as the oxygen demands of the body decline.
The sympathetic nervous system, by contrast, prepares the body for action or, as
some would have it, “the three Fs”—fight, flight, or frolic! The nerves responsible
emerge from the thoracic and upper lumbar segments of the spinal cord (fig. 10-2).
These merge to form a strip of nervous tissue called the sympathetic trunk, which runs
up into the neck and down into the abdomen and sends out branches to reach all re-
gions of the body. The sympathetic nervous system generates many of the symptoms
Figure 10-2. The autonomic nervous system. This has two major divisions: the
sympathetic and the parasympathetic systems. The sympathetic system prepares the
body for action. Nerves from the thoracic and upper lumbar segments of the spinal
cord combine to form the sympathetic trunk, which runs up the side of the vertebral
column. This sends out branches to supply the organs, blood vessels, and glands. Its
actions increase heart rate, open up the bronchioles of the lungs, enlarge the pupils,
direct blood to the muscles and skin, and cause sweating and the rising of hairs in the
skin. The parasympathetic system has the opposite action, preparing the body for a
more passive role. It contributes to many cranial nerves and also to nerves that arise
from the lower (sacral) regions of the spinal cord. It reduces heart rate and makes the
pupils smaller and the bronchioles narrower. It also directs blood toward the internal
organs and promotes digestive activity in the gut.
338 Chapter Ten

of the stress response. Nerves running to adrenal glands cause the release of adrenaline
into the bloodstream. Heart rate increases and blood is directed away from organs such
as the gut and toward the muscles of the limbs and trunk. One of the actions adrenaline
can have is to produce muscle tremor, a problem many musicians are all too familiar
with. The pupils enlarge and the bronchioles in the lungs widen to make it easier to
take in air. The hair on the skin may rise, and we begin to sweat in preparation for
cooling the body when the expected muscle activity starts. In addition, as the parasym-
pathetic system shuts down, the mouth may become dry as the flow of saliva slows.

ANXIETY IN PERFORMANCE

A number of investigations have sought to reveal the physiological consequences of

performance-related anxiety among musicians (Fredrikson and Gunnarsson 1992). In
one study, the musicians were asked to take part in two playing sessions, one of which
was a practice without an audience and the other a performance of the same music in
front of an audience. The musicians were divided into two groups on the basis of how
they themselves rated their typical level of performance anxiety. Interestingly, though
the heart rate during performance was higher in the anxious group, the levels of
adrenaline and cortisol did not appear to differ significantly between the more anxious
and the less anxious musicians. Though many other studies have also reported a link
between heart rate and self-assessed levels of confidence (Abel and Larkin 1990), one
found that though heart rate increases during public performance, it is not related to
subjective anxiety levels (Craske and Craig 1984). Assessment of performance qual-
ity by an adjudicating panel generally confirmed that the anxious students performed
less well. This slightly confusing picture is probably a consequence of a number of
differences between the design of the studies such as 1) the relatively small number of
subjects in each study, 2) possible differences in the levels of experience and perfor-
mance history between the groups of musicians studied, 3) a lack of standardization
in how and when anxiety or confidence were measured, and 4) gender imbalances
between some of the groups compared. This last point may have considerable sig-
nificance as in one study, though the males were found to show greater increases in
blood pressure immediately prior to performance, it was the females who reported
greater levels of anxiety (Abel and Larkin 1990). However, the general trend that
emerges from these studies is that performance anxiety is not related in a simple way
to the physiological manifestations of stress. This emphasizes the importance of trait
anxiety, whether intrinsic or reinforced by experience. We have already encountered
the idea that a degree of stress is required to support high levels of performance, and
this is well recognized by performers (Steptoe 1989). The crucial factor is therefore
how one responds to it. Being inhibited by performance anxiety can be contrasted with
the emotions reflected in phrases such as “being up for it,” which are associated with
a stressful situations where the negative aspects are overridden by positive strategies
used to build self-belief. Consciously or unconsciously, self-congratulatory rituals are
often used by sports teams for this purpose.
 Performance-Related Stress and Its Management 339

COPING WITH STRESS

There is no single magic bullet for dealing with stress. Surveys of coping strategies
that are used by musicians to combat stage fright reveal a broad range of approaches.
These include distraction techniques, the use of ritual, physical strategies including
deep breathing and muscle relaxation, positive self-talk, and the use of drug treat-
ments, preferably under medical supervision (Wolfe 1990; Steptoe 1989). Some tech-
niques are directed primarily at stress management, while for the others this is just
one of a broader range of objectives directed at the physical and emotional well-being
of the performer. Because the origin as well as the nature of the stress response varies
between individuals, so will the attraction and suitability of different coping strate-
gies. There is no right or wrong approach; it is a matter of individual choice. Practical
guidance in some of these techniques can be obtained from a variety of sources (e.g.,
Everly and Rosenfeld 1981; Connolly and Williamon 2004).

Exerting Control and Taking Responsibility

One factor that contributes significantly to how stressful a situation feels is the de-
gree of control that the individual feels he or she can exert over the situation. Jobs
that make high demands yet allow little control over decision making tend to lead to
psychological stress. This is equally true of other life situations such as family rela-
tionships. An investigation into career stress in orchestral musicians suggested that
many felt this to be a significant negative aspect of their working lives, though other
factors such as irregular hours, the monotony of rehearsals, and separation from their
families were also significant (Steptoe 1989). Being or feeling intimidated by the con-
ductor is frequently cited as a major stressor (Wynn Parry 1998). On the other hand,
unlike many stress-inducing occupations, a number of aspects of orchestral musicians’
life provide sources of satisfaction and fulfillment (e.g., job variety, the pleasure of
orchestral playing, and traveling). The long-term effects of control or the lack of it
have been revealed in a series of fascinating studies of civil servants, which have
found that senior grades, who exercise control to the greatest extent, live longer and
have a lower risk of physical ailments such as heart disease than those in more junior
positions (Marmot 1997). One way of reducing stress is therefore to find some way
of exerting control. Some orchestras are now managed by the players, which should
help matters considerably; however, this is not going to be a general solution. If one
feels intimidated at work, taking a course in assertiveness that teaches individuals to
make their feelings or wishes clear in ways that are firm but not based on anger may
make a significant difference. With the increase in confidence this brings may come
a realization that many perceived threats or criticisms are considerably exaggerated.
If an orchestral player does several other things to make ends meet, such as teaching
or session work, it may be that limiting or consolidating these activities will improve
quality of life by reducing stress, even if income declines to some extent. The stress
associated with the job should not be seen in isolation from life’s other stressors such
as balancing family commitments, and assertiveness training may help the performer
340 Chapter Ten

manage these also. The other side of control is taking responsibility for one’s actions
(Everly and Rosenfeld 1981). Each person is to a large extent responsible for much
of the stress in their daily lives in the sense that they themselves make the decision to
take on certain work-related commitments, to adopt a given lifestyle, or to either put
up with or resist particular patterns of behavior in others. A conscious acceptance of
this is the first step in taking control.

Psychological Strategies
One popular approach to the treatment of performance-related anxiety is based on
cognitive-behavioral therapy (Arcier 2000; Connolly 1997; Connolly and Williamon
2004). As the name suggests, this has two aspects, one of which is to change the
subject’s perceptions (cognition) of the stressful situation and hence the emotional
response this engenders, while the second is to control the behavioral response to it.
Stage fright is frequently associated with catastrophizing—being fixated on the worst
possible outcome and its consequences. By contrast, effective coping strategies are
best manifested by a realistic assessment that accepts that while small deficiencies
may occur in a performance, these need not seriously affect audience perceptions and
enjoyment or peer opinion (Steptoe and Fidler 1987). The therapy seeks to promote
the realization that catastrophizing can become self-fulfilling and that in order to break
the cycle it is necessary to replace the negative attitude with a more positive one. On
the other hand, one will want to avoid the opposite extreme of being totally blasé, as
this is unlikely to have a positive effect on performance quality! The behavioral arm
of the therapy involves learning techniques to control the physiological response to
stress. For example, the performer will be taught relaxation techniques and, once in a
calm state of mind and body, will be taught to maintain this while being taken through
a mental exploration of progressively more stressful situations leading up to a major
performance scenario. This combines physical coping strategies (such as awareness
of tension, muscle relaxation, and breath control) that can be used in the run-up to
an actual performance, with an attempt to control the emotional response through
familiarization (Connolly 1997). Of course these, and the purely physical strategies
we will encounter next, also contribute to the feeling of being in control of the situa-
tion because the physical manifestations of stress are kept in check. The therapy may
also involve actual performances, starting with small audiences that are known to be
sympathetic (family, friends, or fellow students) and building up to more typical real-
life audience situations. This may be particularly beneficial where stage fright has
either been triggered initially or subsequently reinforced by bad experiences of public
performance.
Meditation can be an effective means of stress reduction. Though it has been em-
ployed in many forms for centuries, it is only relatively recently that it has gained
wide acceptance and respectability in the West. It may take place within the context
of a religion or philosophy (e.g., Buddhism, certain forms of Christian prayer ritual) or
be an entirely secular device (yoga and transcendental meditation). A central element
of most systems is to focus the attention on a single subject. This can be a rhythmic
physical activity, usually breathing (e.g., paranayama in hatha yoga) or the repetition
 Performance-Related Stress and Its Management 341

of a phrase or syllable (e.g., a mantra such as the Buddhist “Om mani padme hum”).
Alternatively, it may be a mental task such as the contemplation of an irresolvable
problem (e.g., the Zen paradox of one hand clapping) or the visualization of an object
or pattern such as a mandala (Everly and Rosenfeld 1981). Meditation serves to oc-
cupy the mind and reduce or eliminate the attention paid to other thoughts or sensa-
tions. There is growing evidence that the mental states brought about by the different
forms of meditation are basically similar (Newberg and Iversen 2003). They are
characterized by striking changes in the pattern of brain activity, which become more
marked as the skill of the meditator increases (Lutz et al. 2004). In particular, there is
a change in the frequencies of dominant rhythms of electrical activity in the cortex.
The parasympathetic arm of the autonomic system gains the upper hand and metabolic
rate falls. There is also an inhibition of areas of the sensory cortex that provide us
with a three-dimensional awareness of the position of the body in space, and this may
underlie the feeling of physical detachment or of a oneness with the external world
that often results (Newberg and Iversen 2003). The areas of the brain whose roles in
the stress response we have already discussed (the prefrontal cortex, limbic system,
and hypothalamus) are all affected. Meditation tends to lower blood pressure, causing
cells in the hypothalamus to release a hormone called vasopressin from the pituitary.
Vasopressin narrows the arteries to return the blood pressure to normal levels, but it
also appears to reduce feelings of fatigue and to have both a calming effect and a posi-
tive influence on mood (Newberg and Iversen 2003).

Physical Strategies
Some of the physical strategies utilized in stress management have already been en-
countered as elements of more psychological approaches. Two of the most common
are breath control and muscle relaxation (Everly and Rosenfeld 1981). Used on their
own, either of these techniques (but particularly breath control) can be useful during
the period immediately before a performance. Breath control usually involves deep,
slow breathing, which engages the muscles of the abdomen as well as those of the dia-
phragm and chest wall (see chapter 4). The duration of the various phases of breath-
ing (inhalation, holding the lungs full, exhalation) is often regulated by counting. It
is important to avoid overbreathing (breathing deeply and fast when the body is not
active) because this can reduce the levels of carbon dioxide in the blood sufficiently to
induce symptoms such as dizziness, tunnel vision, and even fainting. Though carbon
dioxide is a waste product of respiration, it also plays an important role in controlling
the acid-base balance of the blood. If levels of circulating carbon dioxide fall too far,
the vessels supplying the brain start to shut down, reducing its blood supply. This is
what produces the symptoms just described. Hyperventilation can also be seen as a
symptom of the stress response, for which the simplest remedy is to breathe in and out
from a paper or plastic bag. Rebreathing the exhaled air, which is rich in carbon diox-
ide, is the quickest way to return its concentration in the blood to the correct level.
The muscle tension that is associated with stress can be addressed through relax-
ation exercises. The feature that distinguishes this type of relaxation from simply
slumping into your favorite chair is that attention is directed onto each of the major
342 Chapter Ten

sets of muscles in turn and the tension in them is consciously observed before they
are relaxed. Many systems advocate contracting the muscles strongly before relax-
ing them to make the release more obvious. It is necessary to do this because we
are generally unaware of chronic muscle tension (indeed, if we were aware of it, the
tension would probably never become chronic), so it is quite possible to be lying in
bed thinking you are relaxed, with the muscles in your shoulders knotted and those
of your face tightly contracted. An important element of all relaxation techniques is
therefore to learn to recognize the muscle tension to which you have become oblivi-
ous. Once you are lying or sitting comfortably, muscle groups are worked through
systematically in a simple, easily remembered order. For example, you might start
first with the feet, then the ankles, lower legs, thighs, and so on. Initially one has to
be constantly vigilant against the return of the habitual tension, not only in the large
muscle groups, but also in the smaller ones, particularly those of the face, whose
activity can be highly significant in stress. Apart from the unnecessary tension in
postural muscles that is a feature of stress in everyday life, the musician also needs
to seek out inappropriate tension in the muscles used in performance. Some of these
may also be postural, but others will be those that control the arm or the hand. For
example, in chapter 3, the negative consequences of excessive finger tension and of
the co-contraction of antagonistic finger muscles were discussed. To eliminate this it
is necessary to make a constant and conscious effort to reduce muscle activity to the
minimum that is compatible with playing, something that is much more difficult than
simple relaxation exercises performed outside the context of playing. One approach
that could be more widely employed in treating chronic muscle tension in musicians
would be to use a type of biofeedback in which information on muscle activity is
collected from electrodes placed on the surface of the skin. This can then presented
on a screen to give an objective and real-time indication of levels of muscle tension
(Markison 1998; Levee, Cohen, and Rickles 1976).
Postural therapies such as the Alexander technique, tai chi, or yoga also have a role
to play in stress reduction. These frequently incorporate breath control and muscle
relaxation as part of their regimen but may offer other physical benefits. Follow-
ers of these therapies feel that they derive considerable benefit from them, and this
belief alone may be sufficient to produce positive results. In conventional medicine
nowadays, great emphasis is given to objective testing of new therapies, and it is felt
that the same approach should be used for existing techniques that claim to provide
specific health benefits. One of the difficulties in obtaining evidence of this type is
that large numbers of participants are required by such studies if clear results are to
emerge. Nonetheless, some recent preliminary investigations into the effectiveness
of the Alexander technique on stress in musicians suggest that although it appears to
have no effect on heart rate, it may produce a moderate reduction in levels of anxiety
and an improvement in performance in low-stress situations (Valentine 1995, 2004;
Valentine and Williamon 2003).
The general benefits of exercise and weight control are well known; however, in
the context of stress relief their effect on heart rate and blood pressure is particularly
relevant. Exercise, especially when accompanied by weight loss, reduces both resting
 Performance-Related Stress and Its Management 343

heart rate and blood pressure and the degree to which these increase under conditions
of stress, including those of musical performance (Georgiades et al. 2000; Taylor
and Wasley 2004). Exercise also improves mental outlook or mood, an effect that
becomes more pronounced and long lasting with increasing fitness (Petruzzello, Hall,
and Ekkekakis 2001). In addition, the physical fatigue that results from exercise has a
beneficial effect on sleep patterns. An important aspect of exercise and weight control
through dieting is consistency. Following a strict diet or an intensive exercise program
for a short period and then dropping it will have little or no long-term benefit. Exer-
cise does not need to be an hour of intensive circuit training every day but should be
tailored to your current level of fitness. Choose something that you can easily manage
and hopefully enjoy and build up the intensity as you become able. For example, you
might begin with one session of twenty to thirty minutes once a week for the first
month, then two sessions of similar or slightly greater length for the next month, and
so on (Taylor and Wasley 2004). As your physical condition improves, you will prob-
ably find that you are increasingly looking forward to these sessions and will not want
to miss the feelings of well-being and satisfaction it gives. Exercise can be walking
or cycling to work, swimming, team sports, or aerobics. It might be yoga, tai chi, or
Pilates. It can be a solitary or a communal activity depending on inclination or conve-
nience. You may also want to vary what you do, especially if some of your activities
are likely to be affected by the weather or the season. The exercise sessions should be
built into your daily activities so that they occur within a routine that allocates both to
them and your other commitments adequate time for fulfillment. This will ensure that
the benefits are not immediately dissipated by a stress-inducing rush at the beginning
and end of each session. If you think you do not have time, you may need to reconsider
your priorities! If you are often on tour, you will also want to devise a “portable” form
of exercise such as regular walking or jogging. Some musicians exercise on the day of
a performance as part of their preparation, but before such a scheme is implemented
it is important to experiment carefully in order to work out the best time for this and
how intensive it should be.
Detailed advice on diet is beyond the scope of this book, but the rules for success-
ful weight loss are similar to those for gaining fitness. In the long term, losing weight
means changing eating habits permanently, and this is most likely to be effective
through combining a moderate reduction in food intake with a controlled increase in
exercise. The result of this approach will be a relatively slow but sustained reduction
in weight. It is better to eat a normal healthy diet, but less of it, than to follow one of
the overhyped crash courses based on dubious nutritional premises that are constantly
being publicized. The trouble with such diets is that even if they work, they are often
so grueling that by the end you feel that you deserve the “reward” of returning to old
habits! Instead, you should learn to become accustomed to food or drinks that are less
sweet, to using reduced-fat milk, leaner meat, and unsaturated fats, and to eating more
fruit and vegetables. Diet can also have a more direct influence on the stress response.
Excessive amounts of caffeine can produce symptoms similar to stress and a high salt
intake raises blood pressure, so it is important to control their consumption. A sustain-
able diet is not one without pleasure, but one in which overindulgence is resisted.
344 Chapter Ten

PHARMACOLOGICAL APPROACHES
TO STRESS MANAGEMENT

Drugs of one sort or another have been used in stress reduction for thousands of years.
Some are so familiar to us that we don’t even think of them in this way, but alcohol,
caffeine, and nicotine are as much drugs as tranquilizers, antidepressants, and illegal
drugs of abuse. They all have potent biological effects, and many can be used both for
good and for ill. It is important to recognize the full range of effects that these drugs
can produce and, even if they are prescribed for you by a doctor, to understand their
action and possible side effects. It is up to you to take an active role in the decision
about what is appropriate for you.

Alcohol and Nicotine

The most easily obtainable and traditionally used drugs for stress relief are alcohol and
nicotine, and their use, past and present, has been widespread among musicians. While
moderate drinking may be a perfectly acceptable means of reducing inhibitions and
facilitating social interaction at the postperformance party, it is much less acceptable
as a means of dealing with the anxiety of the performance itself. Even small amounts
will impair the fine motor control needed for playing of a good standard as well as
have an adverse effect on the ability to recall the music, though it may engender a
frame of mind in which this doesn’t appear to matter much at the time! One general
effect of alcohol is to reduce the ability to perceive the loss of physical control, which
makes it difficult for the drinker to find a balance between reducing anxiety and an
unacceptable loss of technique. This will, however, be more obvious to fellow players
and to the audience. Chronic excesses in alcohol consumption may ultimately lead to
an increase in anxiety as well as other psychological problems such as addiction and
physical damage to the liver and other organs. There are many better ways to deal with
performance-related stress than this! It should also be appreciated that even socially
acceptable levels of alcohol intake may be incompatible with other medications for
stress such as tranquilizers and antidepressants. If these are prescribed for you, infor-
mation on their compatibility with alcohol should be sought from your doctor.
Nicotine acts on the nervous system by mimicking one of the chemical neurotrans-
mitters (acetylcholine) that nerve cells use to communicate. Contrary to common
belief, nicotine acts as a stimulant and not a calming agent (Britton et al. 2000). It
increases heart rate and blood pressure and the release of the stress hormone cortisol
into the bloodstream. When the heart works harder, the blood vessels supplying its
muscular walls become wider to increase the oxygen available to the tissue; however,
nicotine can prevent this, especially in those whose arteries are already partly blocked
with fatty deposits. Another effect is to increase hand tremor, which declines only
as the nicotine gradually disappears from the circulation as it is broken down by the
body. Nicotine also increases metabolic rate and suppresses appetite so that weight is
reduced. On ceasing smoking, appetite and weight rise for six to twelve months; how-
ever, both should then return to presmoking levels. Smoking does not reduce anxiety,
and the idea that it does is due to confusion about the origin of the physical symptoms
 Performance-Related Stress and Its Management 345

seen in smokers who have not smoked for a short while or who try to quit. The feel-
ings of irritability and restlessness that ensue are in fact withdrawal symptoms from
nicotine dependence. Though nicotine is highly addictive, it does not in itself cause
cancer. The high incidence of lung cancer among smokers is the result of exposure to
a toxic cocktail of other compounds found in the smoke.

Prescription Drugs
Pharmacological treatment of serious chronic stress in the general population is based
on tranquilizing drugs that reduce anxiety (anxiolytics) or promote sleep (hypnot-
ics) and on antidepressants. We will deal with depression and antidepressants later
in the chapter. The anxiolytics and hypnotics most widely prescribed today belong
to a family of compounds known as benzodiazepines that include such well-known
drugs as Valium (also known as diazepam) and Librium (West 2004). Though they
vary to some extent in their properties, they all produce a degree of sedation, and this
will clearly have a deleterious effect on the ability to carry out the type of complex
behavior that underpins musical performance. In some of the newer benzodiazepines
(e.g., buspirone) these symptoms are less severe; however, in one study of perfor-
mance anxiety that compared buspirone treatment with cognitive-behavioral therapy
in a group of musicians, the drug did not appear to have any marked effect (Clark and
Agras 1991). Though the results must be treated with caution due to the small number
of people involved in the study, there did appear to be a significant improvement in
the group undergoing cognitive-behavioral therapy. Tranquilizers are potentially ad-
dictive and should never be prescribed for long periods. They should only be obtained
on prescription and their use should be closely monitored by a doctor. Given the
consequences of their side effects for the musician, their use in treating stress that is
purely performance related is often questionable (James 1998).
Many musicians have encountered drugs known as beta blockers, which can be
used to cope with the physical symptoms of performance-related anxiety. One much-
quoted and extensive survey of the medical problems in orchestral musicians in the
United States reported that up to 27 percent of the respondees had tried them at one
time or another (Fishbein et al. 1998). Beta blockers act on the tissues that are the
targets of the sympathetic nervous system. Most have relatively little effect on the
brain, so while they reduce the physical symptoms of stress (e.g., increased heart rate
and blood pressure, tremor), they have no direct effect on the psychological aspects of
stress such as the inability to sleep or the presence of negative thoughts. The actions
of the sympathetic nervous system during stress are brought about partly through the
actions of a chemical neurotransmitter called noradrenaline (also known as norepi-
nephrine), which is released from nerve endings, and partly through the release into
the circulation of the hormone adrenaline (or epinephrine) from the adrenal glands.
Noradrenaline and adrenaline bind to receptor proteins (called adrenoreceptors) in the
heart and smooth muscle cells found in the walls of blood vessels or in the visceral
organs to produce the physical manifestations of the stress response. There are sev-
eral types of adrenoreceptors, but the stress response is mainly associated with a type
known as beta-adrenoreceptors. Beta blockers work by interfering with the binding of
346 Chapter Ten

adrenaline and noradrenaline to beta-adrenoreceptors. They were originally developed

for the treatment of angina, where their role is to prevent the heart from becoming
overtaxed during stress and to control high blood pressure, but their effectiveness in
treating stress-related symptoms in normal individuals, particularly musicians, was
quickly realized (Brandfonbrener 1990). There are now many beta blockers available.
These each have slightly different properties; which one is prescribed will depend on
the precise nature of an individual’s stress symptoms (Nies 1990; West 2004). For
example, not all are effective against tremor, which is likely to be an important con-
cern for many players (James 1998). Beta blockers, like all prescription drugs, have
powerful effects. You should therefore never use medication prescribed for someone
else, but should always consult your doctor directly if you are considering these as an
option. He or she will check that they are suitable for you and select the appropriate
type, dosage, and pattern of use. Though quite safe when obtained and used in this
way, they can be dangerous for those with certain medical conditions such as bron-
chial asthma, low blood pressure, and some other problems. In the case of asthma, for
example, they will counteract the effect of the drugs in inhalers, leaving the patient in
a dangerous situation during an attack. They are also not appropriate for performers
who must be physically active, such as dancers, because their effect on cardiac output
will curtail the ability to sustain high levels of muscular activity.
A consideration of the effects of drugs of abuse is beyond the scope of this book,
but an informed and thoughtful consideration of the risks these present, together with
further information on the drugs already mentioned, can be found in West (2004).

DEPRESSION

We all get fed up from time to time and may complain about being depressed. This
mood generally only lasts for a few hours or days at most and we remain perfectly
capable of continuing our normal daily lives. We may respond to feeling down by
indulging in some energetic exercise or by seeking out pleasurable experiences of
one sort or another, for example, taking a holiday or perhaps indulging in “retail
therapy.” Stress can cause this type of depression. For example, it can be triggered
by an unpleasant incident at work that knocks our self-esteem, or the feeling of being
undervalued. Musicians, whose performance is under constant scrutiny not only by
others, but also by themselves, may feel particularly vulnerable. However, prolonged
stress can also lead to a group of altogether more serious conditions that are classified
under the heading of clinical or major depression. There are several types of clinical
depression, but they are usually divided into two broad categories. The first is unipolar
depression, in which the changes in mood are entirely negative. By contrast, bipolar
depression (also known as manic depression) is characterized by violent mood swings
between the extremes of depression and overexcitement.

Unipolar Depression
The main classification scheme used for depression in the United States, the Diagnos-
tic and Statistical Manual of Mental Disorders (DSM-IV; see American Psychiatric
 Performance-Related Stress and Its Management 347

Association 1995) requires the presence of five of the following symptoms over a
period of two weeks for a diagnosis of clinical depression:

• Depressed mood most of the day

• Diminished interest or pleasure
• Significant gain or loss of weight
• Inability to sleep or sleeping too much
• Reduced control over bodily movements
• Fatigue
• Feelings of worthlessness or guilt
• Inability to think or concentrate
• Thoughts of suicide or death

However, such phrases give little idea of what clinical depression actually feels like;
indeed, depressed individuals find it almost impossible to describe their feelings in
words. In severe cases it is an overwhelming sadness and apathy generally coupled
with a total lack of energy so that even simple tasks such as getting out of bed and
taking a shower present insuperable obstacles. This is difficult for the family or friends
of the depressed person to cope with because it represents a mental state so far outside
the bounds of normal experience that it makes it very hard for them to empathize with
it. The natural reaction is to tell the depressed person to “snap out of it,” which is how
we try to respond to a normal bout of the blues, but a request like this to take control
of the situation encapsulates precisely what is impossible for someone suffering from
clinical depression. Clinical depression is a common condition. Approximately 5
percent of the population is likely to be affected at any one time, with women being
twice as likely to experience it as men. The prevalence of depression may come as
a surprise, but even today there is a considerable and completely unjustified stigma
against it, which stops many sufferers from wanting to talk about it. Recently, how-
ever, some very eloquent books have been written about depression by individuals
who have suffered from it. Two of these are Malignant Sadness by Lewis Wolpert
(1999) and An Unquiet Mind by Kay Jamison (1996), and anyone who wishes to un-
derstand more about its various forms and how it can be treated will find these to be
invaluable sources of information.

Bipolar Depression
As the name suggests, a person with bipolar or manic depression experiences abrupt
swings between two extreme emotional states. The depressed state is similar to the one
we have just encountered in unipolar depression. The symptoms of the manic phase
as listed in DSM-IV are given below. The presence of any three is regarded as being
indicative of the condition:

• Overactivity
• Increased talkativeness or pressure of speech
• Flight of ideas or racing thoughts
348 Chapter Ten

• Inflated self-esteem
• Decreased need for sleep
• Distractibility
• Indiscreet behavior with poor judgment
• Marked impairment in occupational or social function

This is probably best illustrated by an example. The writer Virginia Woolf suffered
from bipolar depression, and Wolpert (1999) presents a description that her husband,
Leonard, gave of the manic phase. “She talked almost without stopping for two or
three days, paying no attention to anyone in the room or anything said to her. For
about a day what she said was coherent; the sentences meant something though it
was nearly all wildly insane. Then gradually it became completely incoherent, a mere
jumble of dissociated words.” Though the mania may be characterized by great energy
and resistance to fatigue, the ideas that flood the mind are constantly changing, so
while great enthusiasm may be put into various schemes the individual devises, most
are never carried to fruition. The level of mania is variable, however, and in some
cases a sufficient level of control may remain to allow life to be lived almost normally.
Indeed, it has even been suggested that certain historical figures such as Winston
Churchill, Abraham Lincoln, and Theodore Roosevelt may owe some of their political
success to the prodigious workload that a mild manic state allowed them to maintain
for at least some part of their lives (Snyder 1986).
Many factors may contribute to the onset of depression, and there is still consider-
able controversy about their relative importance. There is undoubtedly a genetic com-
ponent to the susceptibility to depression. In addition, there is evidence that a variety
of social stresses, whether within or outside the family, can be significant contributing
factors. For example, bereavement in childhood or the loss of a supportive partner
are possible triggers, as are the perceptions of constant criticism or stress within
the family setting, either in childhood or adulthood. As in our discussion of general
stress earlier in this chapter, a feature of many of these situations as perceived by the
individual concerned is a lack of control over events and an inability to escape them.
In women, depression may also be associated with pregnancy or childbirth. Many
people suffer from recurrent bouts of depression at more or less regular intervals.
This is sometimes, but by no means always, related to light intensity and day length.
Depression of this type is called seasonal affective disorder and in northern latitudes
is thought to underlie annual mood swings, which are seen to some degree in much of
the population. Treatment is often based on daily exposure to short periods of intense
artificial light, and though this is not effective for all patients, many show significant
mood improvement as a result.
In order to develop treatments for any medical condition, it is necessary to under-
stand not only the external causes but also the organic changes within the body that
underlie it. In the case of depression, current theories and treatments are based on the
belief that the changes in mood are the result of problems in brain chemistry, and in
particular with the activity of one small group of the neurotransmitters that neurons
use to communicate with each other. The main neurotransmitters involved (noradrena-
line and serotonin) appear to modulate the activity of neural circuits; that is, they make
 Performance-Related Stress and Its Management 349

them more or less responsive to stimulation. Intuitively, fluctuations in the effective-

ness of these transmitters appear to offer a logical explanation for changes in mood.
Mania and depression can be seen as extremes of the normal range of moods that we
all experience; however, the level at which the “thermostat” is set differs for each of
us. Those of us who exist closer to the manic end are regarded as extroverts, while
those whose moods lie nearer the depressive end are described as introverts. Individu-
als with manic depression may therefore perceive their transitions between high and
low moods as a movement between two distinct personalities. Normally the activity
of the neurotransmitters responsible for mood is self-regulating so that though our
feelings normally fluctuate, they do so around an average value that is typical of our
personality type. During manic or depressive episodes, the self-regulating mechanism
appears to become locked at one extreme or the other.
As depression appears to be caused by abnormally low levels of noradrenaline or
serotonin in the brain, antidepressant drugs are designed to increase their availability
within the brain. When two nerve cells communicate with each other, a neurotransmitter
released from the axon on one side of the synapse binds to receptor molecules on the
surface of the dendrite that forms the other side to induce some change in its activity.
After a while the transmitter falls off the receptor, but if it is allowed to remain in the
vicinity of the synapse it will soon bind to another receptor, making the size and dura-
tion of the signal between the two neurons difficult to control. In order to prevent this
from happening, the transmitter is removed, usually by pumping it back into the axon
it came from and/or by destroying it with an enzyme. If not enough transmitter is being
released in the first place, blocking either of these two processes will make the small
amount that is available more effective. Antidepressant drugs target both of these pro-
cesses. Those that prevent the transmitters from being pumped back into the axon may
act on both serotonin and noradrenaline (e.g., tricyclic antidepressants) or on serotonin
alone (selective serotonin reuptake inhibitors). Others (the monoamine oxidase inhibi-
tors) block the enzyme that breaks down both transmitters. These are the main classes
of drugs used to treat unipolar depression. There is, however, a paradox in the action of
antidepressants that, despite intensive research, remains unresolved. It is that although
the effects of these drugs on neurotransmitter levels can be seen soon after they are first
given, changes in mood do not appear for an additional four to six weeks. This is just one
indication that our understanding of the mechanism of depression is far from complete
(Castren 2005). Manic (bipolar) depression is usually controlled by the administration
of antidepressants and of Lithium, a mood stabilizer whose precise mechanism of ac-
tion is unknown and whose therapeutic properties were discovered by chance (Snyder
1986). Because of the wide role played by noradrenaline and serotonin, within both the
brain and the autonomic nervous system, antidepressants can have many side effects.
How severe these are is often unpredictable, so finding the most suitable drug is to some
extent a matter of trial and error. In this search, the long time lag between first taking
the drug and any resultant change in mood is a major problem. Furthermore, a treatment
used successfully to treat one bout of depression may not be effective for a subsequent
bout and so the search for a suitable drug must begin all over again.
A number of non-drug-based therapies are also used in the treatment of depres-
sion today. The most notorious of these is electroconvulsive shock therapy (ECT), in
350 Chapter Ten

which a seizure-inducing electric shock is applied to the head. Prior to the procedure
the patient is anesthetized, given a muscle relaxant to prevent overt convulsions, and
ventilated with oxygen. The mechanism of its action is unknown, but its therapeutic
effects appear to result from the induced seizure rather than the shock itself. ECT has
a rather lurid history, but despite this it can prove a highly effective treatment when
used with today’s knowledge and under controlled conditions. It is generally indicated
when drug treatments have proved ineffective, especially if there are suicidal tenden-
cies. The main side effect is some temporary loss of memory.
Another approach to depression is psychotherapy, either used alone or in combi-
nation with drug treatment. The term psychotherapy covers a variety of approaches,
including cognitive and behavioral therapies like those already described, and inter-
personal therapies that concentrate on analyzing relationships and conflicts with other
individuals, which may contribute to the triggering of depression. The evidence that is
currently available suggests that psychotherapy can be as effective as drug treatment
in bringing about a remission from depression (Casacalenda, Perry, and Looper 2002),
though there may be factors in some cases (e.g., suicidal tendencies) that make the
use of antidepressants either strongly advisable or necessary. There is some evidence
that a combination of drug treatment and psychotherapy may be more effective than
either treatment alone and that psychotherapy may help to reduce the chance of recur-
rent episodes of depression, but more extensive and objective studies are needed to
support these claims.

REFERENCES

Abel, J. L., and K. T. Larkin. 1990. Anticipation of performance among musicians: Physiologi-
cal arousal, confidence, and state anxiety. Psychol Music 18:171–82.
American Psychiatric Association. Task Force on DSM-IV. 1995. Diagnostic and statistical
manual of mental disorders (DSM-IV): Iinternational version with ICD-10 codes. 4th ed.
Washington, DC: Author.
Arcier, A. F. 2000. Stage fright. In Medical problems of the instrumental musician, ed. P. C.
Amadio. London: Martin Dunitz.
Brandfonbrener, A. G. 1990. Beta blockers in the treatment of performance anxiety. Med Prob
Perform Artists 5:23–26.
Britton, J., C. Bates, K. Channer, L. Cuthbertson, C. Godfrey, M. Jarvis, and A. McNeill. 2000.
Nicotine addiction in Britain: A report of the tobacco advisory group of the Royal College of
Physicians, ed. Royal College of Physicians of London, Tobacco Advisory Group. London:
Royal College of Physicians of London.
Casacalenda, N., J. C. Perry, and K. Looper. 2002. Remission in major depressive disorder:
A comparison of pharmacotherapy, psychotherapy, and control conditions. Am J Psychiatry
159 (8):1354–60.
Castren, E. 2005. Is mood chemistry? Nat Rev Neurosci 6 (3):241–46.
Clark, D. B., and W. S. Agras. 1991. The assessment and treatment of performance anxiety in
musicians. Am J Psychiatry 148 (5):598–605.
Connolly, C. 1997. Mental skills to optimise musical performance. In Proceedings of the 7th
International Conference on Music Perception and Cognition, ed. J. Renwick. Adelaide,
Australia: Causal Productions.
 Performance-Related Stress and Its Management 351

Connolly, C., and A. Williamon. 2004. Mental skills training. In Musical excellence: Strate-
gies and techniques to enhance performance, ed. A. Williamon. Oxford: Oxford University
Press.
Craske, M. G., and K. D. Craig. 1984. Musical performance anxiety: The three-systems model
and self-efficacy theory. Behav Res Ther 22 (3):267–80.
Everly, G. S., and R. Rosenfeld. 1981. The nature and treatment of the stress response: A
practical guide for clinicians. New York: Plenum Press.
Fishbein, M., S. E. Middlestadt, V. Ottati, S. Straus, and A. Ellis. 1998. Medical problems
among ICSOM musicians: Overview of a national survey. Med Prob Perform Artists 3
(1):1–8.
Fredrikson, M., and R. Gunnarsson. 1992. Psychobiology of stage fright: The effect of public
performance on neuroendocrine, cardiovascular and subjective reactions. Biol Psychol 33
(1):51–61.
Georgiades, A., A. Sherwood, E. C. Gullette, M. A. Babyak, A. Hinderliter, R. Waugh, D.
Tweedy, L. Craighead, R. Bloomer, and J. A. Blumenthal. 2000. Effects of exercise and
weight loss on mental stress-induced cardiovascular responses in individuals with high blood
pressure. Hypertension 36 (2):171–76.
James, I. 1998. Medicines and stage fright. In The musician’s hand: A clinical guide, ed. C. B.
Wynn Parry. London: Martin Dunitz.
Jamison, Kay R. 1996. An unquiet mind: A memoir of moods and madness. London: Picador.
Kendrick, M. J., K. D. Craig, D. M. Lawson, and P. O. Davidson. 1982. Cognitive and behav-
ioral therapy for musical-performance anxiety. J Consult Clin Psychol 50 (3):353–62.
Levee, J. R., M. J. Cohen, and W. H. Rickles. 1976. Electromyographic biofeedback for relief
of tension in the facial and throat muscles of a woodwind musician. Biofeedback Self Regul
1 (1):113–20.
Lovallo, W. R. 2005. Stress and health: Biological and psychological interactions. 2nd ed.
Thousand Oaks, Calif.: Sage Publications.
Lutz, A., L. L. Greischar, N. B. Rawlings, M. Ricard, and R. J. Davidson. 2004. Long-term
meditators self-induce high-amplitude gamma synchrony during mental practice. Proc Natl
Acad Sci USA 101 (46):16369–73.
Markison, R. E. 1998. Adjustment of the musical interface. In The musician’s hand: A clinical
guide, ed. C. B. Wynn Parry. London: Martin Dunitz.
Marmot, M. 1997. Contribution of job control and other risk factors to social variations in
coronary heart disease incidence. Lancet 350:235–39.
Newberg, A. B., and J. Iversen. 2003. The neural basis of the complex mental task of medita-
tion: Neurotransmitter and neurochemical considerations. Med Hypotheses 61 (2):282–91.
Nies, A. S. 1990. Clinical pharmacology of beta-adrenergic blockers. Med Prob Perform Art-
ists 5:27–32.
Petruzzello, S. J., E. E. Hall, and P. Ekkekakis. 2001. Regional brain activation as a biological
marker of affective responsivity to acute exercise: Influence of fitness. Psychophysiology 38
(1):99–106.
Smith, A. M., A. Maragos, and A. van Dyke. 2000. Psychology of the musician. In Medical
problems of the instrumental musician, ed. P. C. Amadio. London: Martin Dunitz.
Snyder, S. H. 1986. Drugs and the brain, Scientific American Library series no. 18. New York:
Scientific American Books.
Steptoe, A. 1989. Stress, coping and stage fright in professional musicians. Psychol Music
17:3–11.
Steptoe, A., and H. Fidler. 1987. Stage fright in orchestral musicians: A study of cognitive and
behavioural strategies in performance anxiety. Br J Psychol 78 (Pt 2):241–49.
352 Chapter Ten

Taylor, A., and D. Wasley. 2004. Physical fitness. In Musical excellence: Strategies and tech-
niques to enhance performance, ed. A. Williamon. Oxford: Oxford University Press.
Valentine, E. R. 1995. The effect of lessons in the Alexander technique on music performance
in high and low stress situations. Psychol Music 23:129–41.
———. 2004. Alexander technique. In Musical excellence: Strategies and techniques to en-
hance performance, ed. A. Williamon. Oxford: Oxford University Press.
Valentine, E. R., and A. Williamon. 2003. Alexander technique and music performance: Evi-
dence for improved “use.” In Proceedings of the Fifth Triennial ESCOM Conference, ed.
C. Wolf. Hannover, Germany.
West, R. 2004. Drugs and musical performance. In Musical excellence strategies and tech-
niques to enhance performance, ed. A. Williamon. Oxford: Oxford University Press.
Wolfe, M. L. 1990. Coping with musical performance anxiety: Problem-focused and emotion-
focused strategies. Med Prob Perform Artists 5:33–36.
Wolpert, L. 1999. Malignant sadness: The anatomy of depression. London: Faber and Faber.
Wynn Parry, C. B. 1998. The musician’s hand and arm pain. In The musician’s hand: A clinical
guide, ed. C. B. Wynn Parry. London: Martin Dunitz.
 Index
References to figures and tables are indicated by italics, and entries in square brackets refer to
material on the accompanying CD-ROM.
abdomen, 18, 27, 32, 104–5, 108, 113–20,
123–24, 166, 183, 308, 324, 336, 341
abdominal wall, 4, 25–26, 104, 106, 108,
113–15, 116, 119, 120–29, 133 abduction,
43, 46, 48, 51, 77, 90
absolute pitch, sense of, 236, 248, 314–16
acoustic bass stop, of organ, 288, 305–6
acromium, 44, 45, 46, 48, 77
Adam’s apple, 145, 179
adduction, 43, 51, 90
adhesive capsulitis, 81
adrenal glands, 184, 336, 338, 345
alcohol, 83, 172, 243, 295, 344
Alexander technique, 40, 69, 73, 84, 342
allergic reactions, 7
allodynia, 15
alto voice (castrato), 186
alveoli, 102, 103, 132
amusia, 316–17
amygdala, 222, 322, 333, 335
anaesthesia, local/topical, 73–74, 78,
161,163, 173–74, 210
anatomical snuff box, 59, 78
androgens, 179,181–85, 223
annulus fibrosus, 18, 19, 20, 23, 28
anterior cingulate gyrus, 323
antidepressant drugs, 344–45, 349–50
antihistamines, 173
anxiety, 14–15, 29, 40, 69, 91, 104, 229,
264, 266, 332, 335, 338, 340, 344–45
353
anxiolytic drugs, 345
aponeurosis, palmer, 79
appoggio, 113
arachnoid, 217
armpit. See axilla
artane, 268
arthritis, 92; rheumatoid arthritis, 78, 80;
osteoarthritis, 46, 80
articulator (voice), 144, 145, 171–72
arytenoid cartilage. See cartilages, of the
larynx
aspirin, 15, 173, 182, 299
Association of British Orchestras (ABO), 66,
299
asthma, 7, 334, 346; and wind playing, 132,
135–36
astrocytes, 216
atemstütze, 113
athetosis, 243
autonomic nervous system, 13, 224, 229,
321, 323, 335–36, 337, 341, 349, auditory
system, 163, 218, 220–21, 227–28, 230,
248–49, 254–55, 257, 276–331
auxiliary muscles of respiration, 108–111
axilla (armpit), 46, 82, 110, 181
axon, 6, 9–13, 81–82, 214, 215–16, 219–20,
224, 238, 241, 288, 302, 332, 249
back pain, 17, 27–29, 38–40, 75
baclofen, 268
354
bagpipes, 73, 79, 321, 326
Baillot, Pierre, 35
baritone voice, 120, 154, 167
basal ganglia, 222, 224, 233, 237, 243, 246,
252, 265, 310, 311–12, 323
basilar membrane, 283, 284, 287
bass voice, 125, 161, 165, 167
bassoon, 78–79, 92, 126, 132, 195, 209, 280
bed rest, 28
Bell’s palsy. See facial nerve palsy
benzodiazepines, 268, 345
beta blockers, 345–46
blood: clotting of, 173, 182; plasma, 102,
104; supply, 2, 5–6, 33, 38, 82, 102, 103,
174, 180, 181, 184, 203, 207, 216–17,
221, 223–24, 230, 233, 265, 293, 295,
317, 336, 338, 341; pH, 103, 341;
pressure, 7, 131, 207, 334, 338, 341–46;
red blood cells, 2, 102; vessels, 5, 64, 71,
82–83, 102, 104, 172–73, 175, 180, 182,
207, 210, 217, 293, 317, 321, 334–35,
337; white blood cells, 2, 175
BOLD (blood oxygen level dependent
signal), 233
bone, 1–4, 8, 13; alveolar, 102, 193, 196
bones, named: carpal, 50–51, 52, 57, 78, 85;
clavicle (collarbone), 8, 23, 24, 44, 45,
46, 63, 64, 83, 106, 111–12, 122, 194;
femur (thigh bone), 25–26; humerus, 33,
44, 45, 46, 48–49, 53, 55, 65, 77, 84,
110–11; hyoid bone, 141, 142, 143, 165–
66; jaw (mandible), 27, 36, 142, 143, 166,
193–94, 196–99, 203, 206, 209, 221, 282;
metacarpal, 50–51, 52, 57; phalangeal
(finger bones), 51, 52, 53, 55, 57, 85;
radius, 44, 50, 51, 52, 53; rib, 26, 64,
83, 102, 111; scapula (shoulder blade),
24–25, 33, 44, 45, 46, 49, 75, 77, 109,
111, 143, 195; sternum (breastbone), 22,
24, 44, 45, 105–6, 111–12, 143, 156, 194;
ulna, 49, 50, 51, 52, 51, 53; zygomatic
(cheek bone), 193
botox, focal dystonia and, 209, 268–70
bow, 35–36, 253; design of, 87, 95, 96
boxwood, 7
Bernouilli effect, 153
Beethoven, Ludwig van, 36, 254, 293, 300,
326
Index
brass instruments, 7, 31, 49, 66, 91, 93, 114,
123, 126, 127, 129, 131, 132–33, 135,
144, 195, 196, 198–99, 200–209, 256,
265, 280, 298–99. See also individual
instruments; ergonomic supports for, 93
brace, for teeth, 196
brachial plexus, 8, 63, 83–84, 93
Braille reading, 265, 269
Brain: cortex of. See cortex; gender
differences in, 230–31, 235, 236, 248–49,
314–16; imaging of, 233–34, 248; nerve
cells of, 213, 214, 215; organisation of,
218, 219, 220, 221, 222, 223–24, 225,
226–31; tumors of, 78, 217, 293
brainstem, 219, 220, 221–23, 238, 241–42
breast bone. See bones, named
breathing: [Animation 4-1 breathing];
[Animation 4-2 breathing, continuous];
body type and, 114–17, 118, 119–20;
posture and, 31, 37, 39, 106, 109–
20, 123–24, 166; in singing, 109,
110–20, 123, 124–25, 126 (see also
plethysmography and Konno-Mead
diagrams); in speech, 115, 117–18, 120,
121; in stress, 341; in wind playing,
123, 125, 126, 12–29, 130, 131–33, 134,
135–36 (see also plethysmography and
Konno-Mead diagrams); quiet, 103, 104,
112, 116, 118, 120, 121
breathy phonation. See phonation, breathy
British Association of Performing Arts
Medicine (BAPAM), xiii, 66, 69
Broca’s area. See cortex, areas of
Broca, Pierre Paul, 312
bronchi, 102
bronchioles, 102, 132, 221, 336, 338
Broschi, Carlo. See Farinelli
burning mouth syndrome, 210
“butterflies” in the stomach, sensation of,
334
bursa, 45, 48–49, 80, [Animation 3-2
shoulder problems]
bursitis, 49, 80, [Animation 3-2 shoulder
problems]
buspirone, 345
Cafferelli, 185
calcium, 2, 184
 Index
carpal tunnel, 59, 61, 64, 65, 82, 85,
[Animation 3-5 wrist position]
carpal tunnel syndrome, 85
carrying, 28–29
carbon dioxide, 102, 132, 341
cartilage, 1–4, 18, 22, 80, 102, 105, 111,
139, 141, 145; articular, 3; hyaline, 3, 4,
6, [Animation 3-1 shoulder]
cartilages of the larynx arytenoid, 148, 149,
150, 155, 157, 160, 162–63, 176, 178,
179, [Animation 5-2 larynx breathing],
[Animation 5-3 larynx, voice quality],
[Animation 5-4 larynx, pitch], [Video 5-1
normal larynx]; cricoid, 140, 145, 148,
149, 150, 160, 163, 179, [Animation 5-2
larynx breathing], [Animation 5-3 larynx,
voice quality], [Animation 5-4 larynx,
pitch]; thyroid, 140, 145, 148, 149, 150,
159, 178, 179, 206, [Animation 5-2 larynx
breathing], [Animation 5-3 larynx, voice
quality], [Animation 5-4 larynx, pitch]
castrati, 185–87
catastrophization, 15, 232, 335, 340
cauda equina, 9
cell body, 9, 12, 214
cello, 13, 37–38, 40, 52, 77, 92–93, 252–53;
spike, 40
central sulcus. See sulcus, central
cerebellum, 218, 219, 220, 231, 236, 237,
238–39, 243, 246, 249, 252, 310, 311
cerebral hemispheres, 219, 228, 247, 249
chairs, 22, 27, 37–39
cheilitis, 7
chest. See thorax
“chills,” sensation of, 323
chin rest (violin/viola), 7, 34–36, [Animation
2-2 violin posture]
chorea, 243
chunking. See memory
Churchill, Winston, 348
cingulate gyrus, 222, 229, 322, 333
circular breathing, 131, 134, 135, 141
clarinet, 7, 92, 126, 131, 195, 203–4, 208,
299; harmonics of, 280; thumb support,
71, 78, 80
clavicle. See bones, named
cleft palate, loss of seal and, 205
coccyx, 18
355
cochlea, 283, 286, 287, 288–91, 297, 299–
300, 302, 304–6
cochlear amplifier, 288, 289, 291, 296
cochlear duct. See scala media
cocobola, 7
cognitive behavioral therapy, 230, 340, 345,
350
cold packs, 71
cold sweat, 229, 334
collar bone. See bones, named
colliculi, 220, 302, 320
colophony, 7
conductors, 39, 77, 243, 300, 332, 339;
sound localization in, 320
coloratura singing, 123, 186
compression. See nerves
compression, as treatment, 71
conchae, 172
consonance, 313, 321
contraceptives, voice and, 173, 183
coracoid process (of shoulder blade), 45, 46,
111
corpus callosum, 218, 222, 224, 229, 231,
236, 247–49, 323
corpus striatum, 323
cortex (of brain), 215, 219–20, 222, 223–30,
291
cortex, areas of: auditory, 225, 228, 255,
291, 300, 302, 303, 306, 307, 308, 309,
310, 311–17, 323, 327; Broca’s area,
237, 310; primary motor, 225, 227, 237,
240, 241–46, 247, 248, 252, 254, 265–66;
orbitofrontal, 322, 323; prefrontal, 312,
335, 341; premotor, 237, 240, 242,
246, 248, 250, 252, 255, 265, 310, 312;
presupplementary motor, 237, 240, 242;
primary sensory (somatosensory), 225,
227, 237, 238–39, 240, 242, 246, 249,
264, 269, 341; sensory association, 230,
237, 240, 249; supplementary motor,
237, 240, 242, 246, 248, 250, 252, 255,
310, 311; visual, 225, 228–29, 230, 242;
Wernicke’s area, 228, 307, 309, 310
Corte, Marchese, 286. See also organ of
Corti
cortical maps: sensory, 226, 227, 239–40,
264–65, 269; motor, 226, 227, 246, 247,
264, 269; auditory, 307, 308
356
cortisol, 336, 338, 344
collar bone. See clavicle
colophany, 7
propolis, 7
cover (of vocal folds), 147, 148, 150, 153,
156, 161, [Animation 5-5 vocal fold
cycle], [Animation 5-6 vocal fold cycle,
continuous], [Video 5-1 normal larynx]
cerebrospinal fluid, 217, 219
cranial nerves, 220, 221, 223–24, 226, 241,
336, 337. See also nerves, named
cricoid cartilage. See cartilages, of the larynx
cubital tunnel, 65, 84, [Animation 3-6 ulnar
nerve]
curse of the MacCrimmons, 79
cushions: support (for guitar), 37; wedge, 38
de Quervain’s syndrome, 78
deafness, 292–93; age-related (presbycusis),
295; conductive, 291; drug induced, 292;
in musicians, 297–300; nerve, 291; noise-
induced, 296, 297
decibel (dB), 278
dendrites, 9, 10, 214, 216
dental problems, and wind playing, 196–97,
203–4
depression, 268, 334, 346–50; bipolar, 347–
49; unipolar, 346–47
dermatitis, contact, 7
dermatome
dermis, 5, 6
Deutsche Gesellschaft für Musikphysiologie
und Musikermedezin, xiii
deviation of wrist, [Animation 3-5 wrist
position]; radial, 50; ulnar, 50, 88, 90
diabetes, 78, 81, 85
diaphragm, 104, 105, 106; role in breathing,
106, 107, 108; role in singing, 113–22,
123, 124, 166; role in wind playing,
128–29
diet, 172–73, 179
didgeridoo: circular breathing and, 135;
vocal tract resonances and, 203
difference tones. See Tartini tones
disks: intervertebral, 18, 19, 20–21, 23,
26–27, 29, 39; prolapsed (“slipped”), 19,
28, 36
dissonance, 313, 316, 321, 323
dorsal horn (of spinal cord), 9, 12
Index
dorsal root ganglion, 9, 12,
double jointedness. See joint hypermobility
double bass, 36–37, 93
DPOAE (distortion product oto-acoustic
emissions), 240
DSM-iv, 346–47
Dupuytren’s contracture, 79, 81
dura, 217
ear, 278, 282, 283, 284–300; external, 282;
middle, 282, 285; inner, 219, 221, 243,
282, 286–91
eardrum (tympanic membrane), 282, 283
earplugs, 299
ear wax, 292–93
ebony, 7
eczema, 7, 334
electroconvulsive shock therapy (ECT), 349
ectomorph, 118, 120
edema. See Reinke’s edema
embouchure, 4, 114, 127–28, 131, 134–35,
193, 194, 195, 196–204; muscles of, 198–
99. See also muscles, named; muscles,
recording activity in
emotional responses to music, 6, 320–21,
322, 323–24
endometrium, 180, 181
endometriosis, 185
endomorph, 117, 118, 119–20
endorphins. See neurotransmitters
enkephalins. See neurotransmitters
epicondyl, 77, 84
epidermis, 5, 6
epiglottis, 139, 140, 143, 145, 146, 149, 202
epilepsy, music and, 326–27
epithelium, 5, 6, 172; of pharynx, 141; of
larynx, 141, 145, 146, 147, 148, 161,
173–74, 176; mucosal, 141
Erdesz, Otto, 95, 96
ergonomics, of instruments, 91–93, 94, 95,
96, 97
esophagus, 125, 139, 140, 141–42, 174,
221–22
estrogen, 180, 181, 182–84, 223
Eustachian tube, 139, 283, 285, 292
exhalation. See expiration
extension, 20, 43
exercise, benefits of, 28, 34, 73–76, 78, 106,
172, 187, 342–43, 346
 Index
exercises, for rehabilitation and mobility,
39–40, 52, 62, 73, 80–81, 84, 86–87, 113,
205, 270
exteroception, 12
expiration, 101, 104, 106, 109, 108, 111,
115, 116, 122, 127–28
expiratory reserve volume, 103, 104, 132
eyesight: and posture, 29; in wind players,
207–8
facial nerve palsy, 210, 285
falsetto register, 153, 155, 156, 157, 158,
159–60
Farinelli, 185–86
false teeth, and wind playing, 204
fascia, deep, 49
Feldenkrais, 40, 69, 73
fiddler’s neck, 6–8
fingers, 4, 12–13, 39, 51, 52, 54, 56, 58; 59–
63, 72–95; bones of (see bones, named;
limitations to movement of); pads of, 6,
12, 73, 239, 251, 263; relaxation of, 11,
29, 36, 342
flexion, 43, 20
flow glottogram, 152, 153, 154, [Animation
5-5 vocal fold cycle], [Animation 5-6
vocal fold cycle, continuous]
flow phonation. See phonation, flow
flute, 7, 31, 32, 33, 49, 85, 87, 93, 94, 126,
128–29, 133–35, 195, 196, 198–99, 202,
269
fMRI (functional magnetic resonance
imaging), 233, 235
focal dystonia, 261–70; brain mapping and,
264–66; of embouchure, 204; history of,
261–62; incidence of, 209, 262; injury
and, 263–64; predisposing factors, 266–
67; symptoms of, 262–63; theories of,
263–66; treatment of, 267–70; of upper
limb, 253, 263–64
forebrain, 218, 219, 222, 223–25, 237, 243,
306, 335
formants, 124, 144, 165–66, 170, 172,
203; singers’, 124, 154, 164, [Video 5-2
singers formant]; tuning, in sopranos,
164, 166, 170; vowels and, 167, 168, 169,
171
French horn, 33, 87, 126, 130, 131, 135, 199,
209
357
frontal lobe (of brain), 218, 224, 225, 306,
309, 311–13, 323
frozen shoulder. See adhesive capsulitis
functional residual capacity, 103, 104, 120,
121
ganglion, dorsal root (of spinal cord), 9, 12
ganglion cysts, 78–79
gastric ulcers, 334
gastroesophageal reflux, 173–76, 180, 334
guitar, 6, 32, 37, 52, 78, 80, 85, 88, 89, 248,
262–63, 269–70; electric, strings of, 8
glands, 174, 180, 292, 321, 336; adrenal,
181, 184, 336, 337, 338, 345; of larynx,
146; pituitary, 180, 218, 222, 224, 321,
336; salivary, 7, 13, 199, 206, 207, 209,
221; sebaceous, 6; sweat, 5, 9,13; thyroid,
151
glandular fever, 334
glenoid, 44, 46
glial cells, 81, 214, 216–17
glottis, 114–15, 124, 129, 133, 139, 145,
148, 149, 151, 153–54, 157, 159–63, 176,
201–2
glottogram. See flow glottogram
glue ear, 292–93
Golgi tendon organs, larynx, 162
Grant, Ulysses S., 316
gray matter, 9, 10, 216, 219, 222, 224–25,
229, 236, 238, 243, 245–46, 249–50, 308,
317
growth hormone, 224
Guercino, 92
Guyon’s canal, 64, 65, 85
gyri (of brain), 227
habituation, 252, 293
halitosis, 174
hair cells, of the inner ear, 286, 287, 288,
289, 290–97, 302; damage to, 291–95,
296
hair follicles, 5, 6, 12
handedness, 52, 229, 231, 246, 249
hand, 4, 31, 42, 49, 51–65, 71, 73–74, 76–
80, 84–86, 88–93, 95, 334, 344; cortical
representation of, 227, 229–30, 239,
241–42, 245–49, 252, 264–65; anatomy
of, 43, 50, 51–56, 57, 58, 59–62, 63–64,
65–67, 72, 89
358
Handel, George Frederic, 185
Haydn, Joseph, 326
hemoglobin, 102, 104, 233
hemorrhage, in vocal folds, 173, 175, 179
hard palate, 140, 141, 205
harmonics, 124, 202–3, 235, 280, 281, 286,
290–91, 297, 302, 304–6, 309, 311–15;
of the voice, 144, 154–55, 157, 164,
165, 167, 170, 182, [Video 5-2 singers
formant]
harp, 34
harmony, perception of, 235, 249, 255, 306,
314
harpsichord, 39
Healthy Orchestra Charter, xii, 66
hearing. See also sound, 276–91; loss of,
291–300 (see also deafness); normal
frequency range of, 279, 280
heart rate, 104, 135, 217, 221, 224, 336,
343–44; stress and, 334, 338, 342
heartburn. See gastroesophageal reflux
herpes, 334
Heschl’s gyrus. See also cortex, auditory,
307
blood pressure, 71, 131, 207, 334, 338,
341–46
hindbrain, 218, 219
hippocampus, 222, 229, 250, 311, 322, 323,
335
histamine, 14
homeostasis, 336
homunculus: motor, 226, 227, 241–42;
sensory, 220, 226, 227, 238
hormone replacement therapy, 183–85
hormones. See also menopause, menstrual
cycle, puberty, 26, 179–86, 223–24;
effect on voice/larynx, 179–86
Horowitz, Vladimir, 253
humerus. See bones, named
Huntingdon’s chorea, 243
hypertension, 334
hyperventilation, 120, 341
hypnotic drugs, 345
hypothalamic pituitary adrenal cortex system
(HPAC), 336
hypothalamus, 218, 222, 223–24, 229, 231,
322, 323, 335–36, 341
hypothenar eminence, 57
hysterectomy, 184–85
Index
ibuprofen, 15, 71, 182
incus. See ossicles
inertive reactance, 153
inflammation, 15, 59, 62, 70–71, 76–78,
80–81, 85, 173, 182, 292
inhalation. See inspiration
injury, performance-related, xii, xiii, 1, 14,
65–67
inspiration, in breathing, 105, 107, 110, 112,
115, 118–21, 132, 134, 135, 341,
inspiratory capacity, 103
inspiratory reserve volume, 103, 109
insula, 323
interaural intensity difference, 318, 319
interaural time delay, 318, 319
interneuron, 9, 10, 13, 214, 215–16, 223, 238
irritable bowel syndrome, 334
ischial tuberosities, 27, 30, 32
isovolume line, 115, 116, [Video 4-3 Konno
Mead sing], [Video 4-4 Konno Mead wind]
jaw. See bones, named
joints, 2–4; capsule of, 3, 29, 46, 49,
78–81, 194, [Animation 3-1 shoulder];
hypermobility of, 79–80; synovial, 3, 23,
46, 49, 59, 76, 78, 80, 85, 193
joints, named: elbow, 4, 32, 39, 49, 50, 53,
55, 63, 64–65, 77–78, 80, 82–84, 90;
intervertebral, 29; interphalangeal, distal
(DIP), 51, 52, 53, 90; interphalangeal,
proximal (PIP), 51, 52, 53, 81;
metacarpophalangeal (knuckle), 51, 52,
53, 58, 90, 92; shoulder, 3–4, 22, 44, 45,
46, 49, 80–81, [Animation 3-1 shoulder];
temporomandibular (jaw), 3, 36, 193;
wrist, 39, 59; zygapophysial, 21
keratin, 5, 196
keyboard, 32, 39, 49–50, 61–62, 66, 77–78,
80, 87–89, 90, 95, 97, 241, 246–48, 251,
254–56, 259, 262–63, 269
Konno-Mead diagram, 115, 116, 117, 119,
127, [Video 4-3 Konno Mead sing],
[Video 4-4 Konno Mead wind]
Kung, Tzu Chen, 326
kyphosis, 18, 22, 26–27, 39
lamina propria (of vocal folds), 146, 147,
148, 150, 173, 177
 Index
language, brain areas associated with, 228,
230, 237, 309, 314, 316
laryngocoele, 205, 206
larynx, 2, 5, 102–3, 108, 124, 139, 140,
141–43, 144, 145–63; breathing and,
[Animation 5-2 larynx, breathing];
control of, 149, 161–63; growth of, 177,
178, 179–80; height of, 112, 124, 142,
143, [Animation 5-1 larynx height];
innervation of, 142, 151, 161; joints of,
160–63, 179; muscles of (see muscles,
named); sensation in 151, 161–63, 174;
structure of, 139, 145, 146, 147–50; wind
playing and, 129, 131, 133–34
laryngitis, 173–74, 176
laryngopharynx, 140
laryngoscope, 131, 145
lentiform nucleus, 237. See also basal
ganglia
learning, of motor skills. See motor learning
librium, 345
lifting, 28–29, 30, 133
limbic system, 222, 229, 266, 322, 335,
341
Lincoln, Abraham, 348
Linburg-Comstock syndrome, 62, [Video 3-1
tendon linkages]
lip, 8, 131, 196, 198–200, 203, 209, 265
lip shield, 197
lips, pressure of mouthpiece on, 201
ligaments, 3; annular, 49; ligamentum
flavum, 22; nuchal, 23, 25, 44; vocal,
146, 147, 148, 149, 150, 153, 157, 158,
159, 177, 178
lithium, 349
long bones, 2, 186
longbowmen, 2
lordosis, 18, 22, 26–27, 30, 32, 37–39,
[Animation 2-1 stand sit]
loss of seal, 135, 204–5, 208
lumbar support, 27, 37–38
lung pressure: in singing, 115, 120; in wind
playing, 132–33, 207
lungs, 5, 102–5, 115, 120, 124–25, 127–29,
132–33, 135, 184, 207, 345
magnetoencephalography (MEG), 233
Majorano, Gaetano. See Cafferelli
malleus. See ossicles
359
mandible. See bones, named
manic depression. See depression, bipolar
Manzuoli, Giovanni, 185
Mary Rose, Tudor warship, 2
marrow, 2
Mavron quartet, ix, xv, xvii, 293
Médecine des Arts, xiii
meditation, stress and, 340–41
medulla (oblongata), 218, 219, 220
Meissner’s corpuscle, 6, 12,
melody, perception of, 235, 249–50, 255,
309, 310, 311–12, 314, 316–17, 320
memory, 87, 163, 228, 234, 252, 254, 256,
258, 259–60, 268, 310, 311, 314, 317–18,
355; associative, 229, 250, 335; auditory,
257; chunking in, 257; emotional coloring
of, 322, 355; lapses of, 257, 334, 350;
motor, 13, 242, 252, 257, 266; of the
score, 251, 253, 257, 260; working, 311–
12, 315, 334
meninges, 217, 219
Menière’s disease, 293, 295
menopause, 2, 26, 183–85, 187, 292
menstrual cycle, 181, 213, 316; voice and,
180–81, 182, 183–84
mental rehearsal, 71, 87, 244, 253–55
Merkel ending, 6, 12
mesolimbic pathway, 322–23
mesomorph, 117, 118, 120
Messaien, Olivier, 318
Meyerbeer, Giacomo, 185
mezzo-soprano voice, 125, 164, 167, 186
microglia, 217
midbrain, 218, 219, 220, 224, 243, 302, 322,
323
migraine, 334
mirror neurons, 228, 312
missing fundamental illusion, 304–6, 308,
313, [Sound Presentation SP9-3 missing
fundamentals]
modal register, 155, 156, 157, 158, 159
Modigliani figure, voice and, 184
modiolis, of embouchure, 199
mononucleosis. See glandular fever
Moreschi, Alessandro, 185–87
motor neuron, 9, 10, 11, 13, 210, 215, 241,
245, 268,
motor neurons, synchrony of, 61
motor pool, 10, 241–42, 245
360 Index
motor skill, acquisition of, 237, 240, 243,
244, 246, 251–52, 258
motor unit, 10, 11, 241
mouthpiece, 7–8, 127, 131, 144–45, 195,
196–97, 199–201, 203, 308, 280, 299
“Mozart effect,” 324–26
Mozart, Leopold, 185
Mozart, Wolfgang Amadeus, 185, 245,
324–26
MRI (magnetic resonance imaging), 233,
235, 245, 248
mucosa. See epithelium, mucosal
mucosal wave (of vocal fold), 152, 153,
[Animation 5-5 vocal fold cycle], [Video
5-1 normal larynx]
muscle, 4–6, 10, 12, 14; cardiac, 5;
composition of, 4; fast fibers, 4; fibers of,
4, 10, 11; slow fibers, 4; smooth, 5,13;
striated, 4, 11
muscle wasting, 65, 85
muscles, co-contraction of, 91, 114, 261,
263, 342
muscles, named, of back and neck: erector
spinae, 20, 23, 24–26, 28–29, 30, 39;
longus capitus, 23, 24; longus coli, 23,
24–25; quadratus lumborum, 23, 26;
rectus abdominus, 23, 25–26, 29, 32, 105,
106, 108, 109; psoas, 23, 25; scalenes,
33, 83, 93, 110, 111–12, 195; splenius
capitus, 23, 24; sternocleidomastoid,
23, 24, 33, 35, 110, 111–12, 194, 206;
transversospinalis, 23, 24
muscles, named, of ear: stapedius, 283, 284,
285, 295; tensor tympanum, 282, 283
muscles, named, of embouchure, jaw, and
palate, 194; buccinator, 135, 198–99, 194;
depressor anguli oris, 199, 200, [fig 6-1
latin]; depressor labii inferioris, 199, [fig
6-1 latin]; levator anguli oris, 199, 200,
[fig 6-1 latin]; levator labii superioris,
198, [fig 6-1 latin]; orbicularis oris,
198–200, [fig 6-1 latin]; orbicularis oris,
rupture of (see Satchmo’s syndrome);
masseter, 36, 193, 194, 197–99 [fig 6-1
latin]; platysma, 199, 200, [fig 6-1 latin];
pterygoids, 194, 198; risorius, 199, [fig
6-1 latin]; levator palati, 141, 204; tensor
palati, 141, 204; temporalis, 36, 193, 194,
197–99, [fig 6-1 latin]
muscles, named, of larynx (or acting upon
it): cricoarytenoid, lateral, 124, 148,
149, 157, 160, [Animation 5-2 larynx
breathing], [Animation 5-3 larynx,
voice quality]; cricoarytenoid, posterior,
148, 149, 151, 155, 158, [Animation
5-2 larynx breathing], [Animation 5-
3 larynx, voice quality]; cricothyroid,
124, 149, 150–51, 156, 157, 158,
159–60, [Animation 5-4 larynx, pitch];
interarytenoid, 148, 149, 155, 157,
162, 179, [Animation 5-3 larynx, voice
quality]; omohyoid, 142, 143, 166;
sternohyoid, 142, 143, 165; sternothyroid,
142, 143, 165; thyroarytenoid, 142, 146,
148, 149, 150, 156, 157, 159–60, 162,
[Animation 5-4 larynx, pitch]; thyrohyoid,
142, 143, 165; transverse arytenoid, 124,
146, 149, 150, 153, 159–60, [Animation
5-7 vocal timbre] (see also interarytenoid;
vocalis)
muscles, named, of lower limb: gluteus
maximus, 20, 26, 30, 112; hamstrings, 3,
24, 26; psoas, 23, 25; quadratus femorus,
30
muscles, named, of respiration: auxiliary,
108–13; diaphragm, 102, 103, 104–8,
109, 113–15, 117, 119–24, 128–29, 133,
166, 341, [Animation 4-1 breathing],
[Animation 4-2 breathing, continuous]
(see also diaphragm); intercostals, 102,
105, 106–8, 113–15, 120, 122, 166, 341;
intercostal, external, 105, 106, 109, 115;
intercostal, internal, 105, 108, 109; lateral
abdominal, 105, 106, 108, 109, 125,
[Animation 4-1 breathing], [Animation 4-
2 breathing, continuous] (see also internal
and external oblique and transverse
abdominal); latissimus dorsi, 47, 110;
oblique, external, 106, 125; oblique,
internal, 106, 125; pectoralis major, 33,
50, 110, 101; pectoralis minor, 46, 48,
110, 101; rectus abdominus, 20, 25–26,
29, 30, 105, 106, 108, 110; scalenes,
33, 83, 93, 110, 111–12, 195; serratus
posterior inferior, 110; serratus posterior
superior, 110; sternocleidomastoid,
23, 24, 33, 35, 110, 111–12, 194, 206;
subclavius, 109, 110, 111; transverse
 Index
abdominal, 106; trapezius, 23–25, 44, 47,
48, 110, 111–12, 206
muscles, named, of upper limb and shoulder:
abductor of thumb (long), 55, 56, 57, 59,
78; abductor of thumb (short), 54, 57,
59, 78; adductor of thumb, 59, 63, 65;
biceps, 4, 44, 49, 50, 65; brachilis, 49, 50;
brachioradialis, 49, 50; coracobrachialis,
49, 50; deltoid, 33, 35, 45, 46, 47, 48,
49, 80; extensors of the wrist, 50, 55;
extensors of the fingers, 55, 56, 57, 61,
77, 91, 263, 267; extensors of the thumb,
55, 56, 57, 59, 78; flexors of the wrist,
50, 53, 64, 84; flexors of the fingers, 53,
54, 55, 57, 58, 61–62, 65, 90–91, 263,
167; flexors of the thumb, 53, 54, 57,
62; infraspinatus, 45, 46, 47 (see also
rotator cuff); interosseus muscles, 54, 56,
57, 58, 62, 64, 65, 77; latissimus dorsi,
47, 110; levator scapulae, 24, 33, 44, 47;
lumbricals, 10, 54, 57, 58, 62, 64, 65, 77,
[Animation 3-3 lumbricals]; opponens, of
little finger and thumb, 57, 59; pectoralis
major, 33, 50, 110, 111; pectoralis
minor, 46, 48, 110, 111; pronator teres
and quadratus, 49, 51, 65; rhomboids,
24, 44, 47; serratus anterior, 46, 47, 48;
subscapularis, 46, 48 (see also rotator
cuff); supinator, 51; supraspinatus, 33–34,
45, 46, 48, 49, 77, 80, [Animation 3-2
shoulder problems] (see also rotator cuff);
teres major, 48; teres minor, 46, 48 (see
also rotator cuff); trapezius, 23–25, 44,
47, 48, 110, 111–12, 206; triceps, 49, 50,
64–65, 245
muscles, named, of vocal tract and mouth:
constrictors of the pharynx, 141, 143,
198; digastric, 142; genioglossus, 143,
142; geniohyoid, 142; hyoglossus, 142,
143; myelohyoid, 142; palatoglossus,
143; levator palati, 141, 204; styloglossus,
143; stylohyoid, 142; stylopharyngeus,
141, 142, 143, 166; tensor palati, 141,
204
muscle spindles, of larynx, 162
muscle tension, 25, 27, 29, 36, 39, 40, 44,
69, 86–88, 94, 111, 120, 131, 196, 198,
267, 340–42
musici. See castrati
361
Musician’s Benevolent Fund (MBF), xiii, 66
Mustafa, 185
myelin, 81–82, 214, 216, 236, 248
naked sensory ending, 6, 12
Napoleon Bonaparte, 186, 206
nasopharynx, 141, 171, 204
neck, 24–25; breathing and, 106, 111–12;
muscles of. (see muscles, named); nerves
of, 64, 83–84, 222, 241, 336; vertebrae
of, 20, 21, 23, 44, 109; posture and, 31,
32, 34–35, 37, 66, 86; structure related
to voice, 140, 151, 165–66, 179, 194;
tension in, 27, 29, 36–37, 39, 111; wind
playing and, 195, 205–6; dystonia and,
261
nerve root, 18, 19, 128, 35, 64, 83
nerves: compression of, 8, 13, 19, 33, 63,
64, 65, 69, 78, 81–85, 207, 210, 263–64,
267, [Animation 3-6 ulnar nerve]; axons,
regrowth in, 13, 82, 210; cranial, 202,
220, 221, 223–24, 226, 241, 302, 336,
337; double crush of, 82; segmental, 8, 9,
10, 18; structure of, 81–82
nerves, named: abducent, 220; auditory,
221, 287, 241, 303, 313 (see also
vestibulocochlear); digital, 85; facial,
198–99, 209–10, 220, 221, 226, 285;
glossopharyngeal, 220, 221, 226;
hypoglossal, 221, 220, 226; laryngeal
(superior and recurrent), 142, 151;
median, 63, 64, 65, 83, 85, 89, 264;
oculomotor, 220, 221; olfactory, 221;
optic, 207, 221; radial, 64–65, 72;
trigeminal, 198, 210, 220, 221, 226, 282;
trochlear, 220, 221; ulnar, 63, 64, 65, 82–
85, 263–64, [Animation 3-6 ulnar nerve];
vestibulocochlear, 220, 221, 302 (see also
auditory); vagus, 220, 221, 222, 226
nerve entrapment. See nerve compression
neurotransmitter: acetylcholine, 344, 268,
270; adrenalin (epinephrine), 333, 336,
338, 345; dopamine, 268, 322–23;
endorphin, 15, 323; enkephalin, 15,
323; CGRP, 14; glutamate, 14, 287;
noradrenalin (norepinephrin), 345,
348–49; serotonin, 348–49; substance P,
14, 15
nickel, 7–8
362
nicotine, 344–45
19-norsteroids, 183
noble posture, in singing, 120
nucleus accumbens, 322, 323
nucleus pulposus, 19, 20–21, 27–28, 39
oboe, 7, 92, 103, 126, 130, 131, 195, 202,
204, 207–8, 269, 280, 297
occipital lobe (of brain), 218, 224, 225, 227
oligodendrocytes, 216
omega sign (in primary motor cortex), 247
opposition, 43, 51, 164
organ of Corti, 286, 287, 288, 290–91, 295,
296
organ pipes, 280, 288, 290–91, 295
oropharynx, 140, 141, 172
Ositrakh, David, 36
ossicles of ear, 282, 283, 285, 291–93,
300; malleus, 282, 283; incus, 282, 283;
stapes, 282, 283, 284, 285, 292
osteoporosis, 2, 26, 184
otitis media, 292
otoacoustic emissions, 288, 290
otosclerosis, 292–93, 295, 300
oval window, 282, 283, 286, 292
overbreathing. See hyperventilation
overbite, of teeth, 196, 197, 203
overjet, of teeth, 196, 197
overuse syndrome, 74–76
oxygen, 2, 5, 82, 102–4, 120, 180, 216–17,
233, 285, 295
Pacinian corpuscle, 6, 12
Paganini, Niccolò, 79
palate, 139, 143, 144, 145; soft, 134, 135,
140, 141, 144, 171, 204–5, 221
pain, 12, 13–16, 22, 36, 2, 46, 49, 62, 65,
67, 70–71, 80, 83, 105, 111, 161, 174,
182, 185, 196, 208–10, 215, 263, 278,
334; allodynia, 15; burning mouth, 210;
back, 17, 27–29, 38–40, 75; chronic,
15–16, 28–29, 33, 76; gate theory of, 15;
hyperalgesia, 15; windup, 15
painkillers, 15, 182; aspirin, 15, 173, 182,
292; ibuprofen, 15, 71, 182; nonsteroidal
anti-inflammatories (NSAIDs), 15;
opiates, 15
parietal lobe (of brain), 218, 224, 225, 227–
28, 242, 249, 309
Index
paraphenylenediamine, 7
parasympathetic nervous system, 336, 337,
338, 341
Parkinson’s disease, 224, 243, 268, 312
pear-shaped down posture, 119, 120, 166
pear-shaped up posture, 118, 119, 120, 123,
166
perfect pitch. See absolute pitch, 314–16
Performing Arts Medicine Association
(PAMA), xiii, 70
peripheral nervous system, 8, 12, 13–15
phalangeal cells, 289
phalangeal bones (phalanges), 51, 52, 53,
55, 57
pharyngocoele, 205, 206
pharynx, 114, 125, 131, 133–34, 139, 141,
143, 144, 145, 151, 165–67, 169, 171–72,
174, 179, 193–94, 198, 201, 204–6, 221,
227
phonation: breathy, 153, [Animation 5-3
larynx, voice quality]; flow, 124, 155,
[Animation 5-3 larynx, voice quality];
pressed, 124, 155, 176–77, [Animation 5-
3 larynx, voice quality]
phrenology, 234
physiotherapy, 69–70, 72–73, 78–79, 84
pia, 217
piano, 29, 31, 32, 33, 39, 50, 57, 62, 79, 85,
88–89, 90, 91–92, 95, 161, 241, 246, 247,
248, 252–54, 256–57, 260, 262–63, 266,
269–70, 279, 280, 302, 315–16, 320, 324,
[Animation 2-3 piano posture]. See also
keyboard
piano playing, analysis of, 89–90, 256, 258,
259–60
piccolo, 130, 195
Pilates, 40, 106, 343
piriform recess, 165–66
pitch: absolute, 253, 255, 280, 300, 301,
302, 304–6; 308, 314. See absolute
pitch, sense of; amusia; relative pitch,
sense of
pitch contour, 306, 310
pituitary gland, 180, 218, 222, 224, 336, 337,
341
planum polare, 309, 310, 311
planum temporale, 236, 306, 307, 310, 311,
313, 315, 318, 320
pleasure, pathways mediating, 322, 323–24
 Index
plethysmography, 114–15, 121, [Video 4-
1 plethysmography sing], [Video 4-2
plethysmography wind]
pleura, 102, 105–6
pleurisy, 105
pneumoparotitis, 206, 207
pons, 218, 219, 220
posture: in instrumental playing, 17, 22,
24, 28, 29–31, 32, 22–39, 59, 61,
67–68, 76–78, 83–84, 87–88, 89, 90,
91–93, [Animation 2-2 violin posture],
[Animation 2-3 piano posture]; in singing,
106, 109–10, 112, 117–18, 120, 123–24,
166. Refer also to individual instruments
postural training and exercises, 36, 39–40,
342
practice, 13, 186, 196–97, 236, 248–51, 264;
deliberate, 240, 236, 259–60; learning
and, 242, 251–52, 255–57, 259–60;
mental, 29, 37, 40, 75–77, 86–87, 260–61
(see mental rehearsal)
presbycusis. See deafness, age-related
pressed phonation. See phonation, pressed
pressure: in airway (singers), 143, 151,
153–54, 161–63; in airway (wind
players), 126, 130, 199, 200, 202, 204–8;
of mouthpiece, 196, 200–201; on teeth,
196, 203–4
progesterone, 180, 181, 182–84
Prinner, Johann Jacob, 92
pronation, 35, 43, 49, 51, 65, 95, 96
propolis, 7
proprioceptor, 12–13
prehensile grip (hand), 57
prephonatory posturing, 115, 116, 120, 121
pressed phonation. See phonation, pressed
puberty, the voice and, 177, 179–82, 186
pulse register, 115, 156, 157
pyramidal tract, 241
Quantz, Joachim, 186
Rachmaninov, Serge, 79, 254
radius, 49, 50, 51, 53
Ramazzini, Bernardini, 66, 261
Ravel, Maurice, 317
recorder, 114
reed, 103, 108, 114–15, 127, 129, 132–35,
141, 144–45, 196–97, 199, 203
363
reed instruments, 132–35, 196, 199, 202, 204
reflexes, 13, 134, 220, 245, 285, 302; of
larynx, 162–63; stapedius, 285, 295
register. See vocal register
Reinke’s edema, 173–75, 182, [Video 5-3
edema nodules]
relative pitch, sense of, 314–15
relaxation line, 108, 109, 115, 116, [Video 4-
3 Konno Mead sing], [Video 4-4 Konno
Mead wind]
relaxation, stress and, 339–42
repetitive strain injury, 74, 76, 263
residual volume, 103, 104
resonator (voice), 144, 145, 165–70, 171
respiratory braking, 108, 109, 122, 127–29
respiratory movements, 105, 107
rest (as treatment), 28, 70–71
retinaculum, [Animation 3-4 retinacula of
wrist]; extensor, 56; flexor, 54, 59, 64,
65, 79, 85
rhythm, perception of, 235, 249–50, 306,
308–9, 310, 311–12, 316, 326
rib cage, 8, 20, 46, 104, 111, 117, 120, 123,
127–28, 204
rib movements, in breathing, 105–6
rima glottis. See also glottis, 145
Rimsky-Korsakoff, Nikolai, 318
ring splints, 72, 78, 80
ring, vocal, 124, 167
Rivinus, David, 95, 96
Roosevelt, Theodore, 348
rosewood, 7
rosin, 7
Rossini, Gioachino, 185
rotator cuff, 46, 47, 48, [Animation 3-1
shoulder]; syndrome, 34, 36, 68, 77, 81,
[Animation 3-2 shoulder problems]
round window, 282, 283
Rubinesque figure, voice and, 184
Rubenstein, Arthur, 253
Ruffini ending, 6, 12
Sacks, Oliver, 229
salivary gland, 7, 13, 141, 199, 206, 209, 221
“Saturday night” palsy, 82
Satchmo’s syndrome, 206, 208–9
saxophone, 92, 132, 202–3, 208, 280
scala media, 286, 287
scala tympani, 286, 287
364
scala vestibuli, 286, 287
scapula (shoulder blade), 24–25, 33, 44, 45,
49, 75, 77, 109, 111, 143, 195
Schumann, Clara, 256
Schumann, Robert, 293
Scriabin, Alexander, 318
Score: learning, 71, 87, 253, 255–57 (see
also memory); reading, 221, 250, 253–54
scoliosis, 22, 34
scrivener’s palsy, 261 (see also focal
dystonia)
Schwann cells, 216
seasonal affective disorder, 348
seating, posture and, 37–39, [Animation 2-1
stand sit]
semicircular canals, 283
sensory neuron, 9, 10–14, 215, 243, 245
setup, instrumental, 7, 36, 40
Shakespeare, William, 326
Shebalin, Vissarion, 317
Shostakovitch, Dimitri, 317
shoulder blade. See bones, named, scapula
shoulder rest, 8, 35, 95, [Animation 2-2
violin posture]
shoulder joint. See joints, named
skill learning. See motor skill, acquisition of
skin, 4–5, 6, 7–8, 64, 65, 73, 78, 81, 215,
238–39, 243, 245, 263, 306, 337;
sensation in, 6, 11–14, 73, 183, 186, 198,
224, 268; callus on, 6; reaction to metals,
7. See also dermatitis, contact
slipped disk. See disk, prolapsed
Smetana, Bedřich, ix, xv, 293, 294, [Smetana
quartet, Mavron.wma]
smoking, 127, 173, 344. See also nicotine
soft palate. See palate, soft
somatization, 68, 334
somatotype, 83, 91; breathing and, 117, 128
soprano voice, 124–25, 161, 164, 166–67,
169, 170, 179, 186
sound: frequency, coding in the nervous
system, 286, 300, 301, 302, 304–6,
307; harmonic structure of, 280, 281,
282, [Sound Presentation SP-1 sound
identification]; intensity, 278; localization
of, 318, 319, 320; nature of, 276, 277,
278–82; occupational exposure limits,
299; pressure, 278
Sound Ear Initiative, xii, 298
Index
speech, breathing in, 117, 118, 120, 121
Spohr, Louis, 34
spine (vertebral column), 8–9, 17, 18, 19,
20, 21–22, 23, 24–31, 34–35, 37, 39, 44,
83, 105–6, 111, 139, 150, 184, 195, 217,
337
spinal cord, 1, 8, 9, 10–11, 12, 13–15, 18,
22, 28, 64, 76, 81, 217, 219–20, 223–24,
237, 238, 240, 241–45, 336
splinting, 71, 72, 81
selective serotonin re-uptake inhibitors, 349
stage fright, 332, 339–40
standing, 22, 20, 26–27, 29, 31, 36, 106,
112, 122, [Animation 2-1 stand sit]
stapes. See ossicles of ear
stapedius reflex, 283, 284, 285, 295
state anxiety, 332
stereocilia, 286–88, 289, 290, 295, 296
steroid hormones, 26. See also androgens
and testosterone
sternum. See bones, named
stress, 13, 25, 27, 36, 36, 44, 68, 69, 76, 91,
111, 135–36, 172, 205, 210, 229, 257,
266, 295, 332, 333, 334–38, 348; acute,
332; chronic, 332; management of, 36,
40, 67, 339–46
stretching, 37, 39–40, 52, 76, 80–81, 86–87
stringed instrument, 6, 13, 31, 34, 36–37,
39, 44, 46, 52, 57, 61, 66, 73, 77, 79–80,
85, 87–88, 91, 93, 95, 132, 236, 239,
248, 250, 253, 262–63, 269, 279–80, 282.
Refer also to individual instruments
string playing, analysis of, 88, 25–26,
252–53
“strohbass” register. See pulse register
strokes, 230, 233, 261, 265, 270, 317
stool, 36–37; kneeling, 32, 38; piano, 32, 39,
88, [Animation 2-3 piano posture]
styloid process, 142
subglottal pressure, 122–24, 153–54, 163
substantia nigra, 224
sulcus, 224–25, 234, 239; central, 218, 225,
227, 238, 241, 246, 247
superior olivary nucleus, 291, 302, 303
superior temporal gyrus, 311–13, 317
supination, 43, 49, 51, 77, 93, 94
support, in singing, 110, 112–15, 124–25,
177, 183
supraglenoid tubercle, 49
 Index
surgery, 62, 67, 70, 73–74, 78–79, 81, 83–
85, 151, 175–76, 205, 241, 267, 292–93,
317
swallowing, 139, 141–43, 145, 174, 205,
221–22, 285, [Animation 5-1 larynx
height]
sweat, 5, 17, 86, 224, 295, 338; cold, 229,
334; gland, 5, 6, 13, 321, 336
sympathetic nervous system, 336, 337, 345
synesthesia, 318
synapse, 214, 215–16, 349
syncope, in wind players, 207
synovial fluid, 3, 23, 46, 59, 80
tai chi, 40, 69, 73, 342–43
Tartini, Giuseppe, 288
Tartini tones, 288, 290, 304, [Sound
presentation SP9-2 Tartini tones]
taxi drivers, memory of, 250
tectorial membrane, 286, 287, 295–96
teeth: grinding of, 36, 91; wind playing and,
193, 195, 196–98, 203–4, 210, 299
tempo, 256, 311–12, 314, 324–25
temporal lobe (of brain), 218, 224, 225,
227–29, 250, 282, 306, 307, 308–9, 310,
311–12, 323
tendon, 2, 3, 4–5, 12, 26, 49, 60, 62,
70–80, 85, 88, 219, 238–39, 243,
245; inflammation of, 55, 56, 57,
61–62 (see also tendonitis); sheath (see
tenosynovium; trigger finger)
tendons, named: biceps, 49; extensor (of
fingers), 55, 56, 57, [Animation 3-3
lumbricals], [Animation 3-4 retinacula of
wrist]; extensor expansion (of fingers),
55, 58, [Animation 3-3 lumbricals];
flexor, deep (of fingers), 53, 54, 55,
57, 90, [Animation 3-3 lumbricals],
[Animation 3-4 retinacula of wrist];
flexor, superficial (of fingers), 53, 54,
[Animation 3-3 lumbricals]; interosseus,
57; lumbrical, 45, 57; of long muscles
of the thumb, 57, 59, 78; supraspinatus,
34, 46, 48, 77, [Animation 3-2 shoulder
problems]
tendonitis, 76–79
tenor voice, 125, 161, 165
tenosynovia, 54, 56, 60, 61, 76, 78
tenosynovitis, 76–78
365
tension. See muscle tension
testosterone, 179, 183, 186, 229, 235
total lung capacity, 103, 104, 129, 132
temporomandibular joint syndrome, 36
thalamus, 218, 222, 223, 229, 237, 238, 243,
322
thenar eminence, 59, 65, 85
thoracic outlet, 63, 64, 83; syndrome, 64,
83–84
thoracic recoil, 108, 114, 120, 122, 128–29,
[Animation 4-1 breathing], [Animation 4-
2 breathing, continuous]
thorax (chest), 8, 22, 25, 31, 39, 44, 46,
47, 75, 92, 102, 104, 105, 106–15, 116,
117–18, 119, 120, 122–23, 125, 136, 179,
204–5, 207, 306, 334, 341
throat infection. See laryngitis
throat singing, 169
threshold shift, temporary (of hearing), 296,
298
thyroid cartilage. See cartilages, of the larynx
thyroid notch, 145, 150
tics, 243
tidal breathing, 103, 104, 117
tidal volume, 103, 104
timbre, 235, 280, 302, 305–6; perception of,
255, 310, 311–12; vocal, 144, 154–55,
157, 159–60; wind playing, in, 128,
202–3
tinnitus, xv, 292–93, 294, 295–96, 300,
[Smetana quartet, Mavron.wma]
tip links, of hair cells, 287, 290, 295
tone deafness. See amusia
tongue, 124, 131, 134, 133–35, 139, 140,
141, 142, 143, 144, 145, 166, 169, 199–
203, 214, 221, 227, 238; vowels and, 168,
169–70, 171
tonsils, 140, 179, 205
tonotopy, 284, 286, 300, 301, 302, 303,
305–6, 308
trachea, 102, 104, 106, 139, 140, 141–42,
145, 146, 149, 150, 161, 166, 170, 175,
205, 223
trait anxiety, 332, 338
treatment, forms of, xii–xiii, 13–16, 28, 36,
66–67, 68, 69–74, 76, 78–81, 84–85,
175–76, 182, 185, 196, 210, 264, 267–70,
292–93, 295, 334, 340, 345–46, 348–50
tricyclic antidepressants, 349
366 Index
trigger finger, 60, 78
trigeminal neuralgia, 210
trillo, in Renaissance music, 123, 129
trombone, 33, 126, 131, 166
trumpet, 33, 49, 126, 131, 135, 145, 166,
198, 200–202, 204, 207–9, 297–98
tympanic membrane. See eardrum
tuba, 126, 129, 131, 132, 202
turbinate bones, 172
Tuva, 170
ulna. See bones, named
underbite (of teeth), 196, 197
underjet (of teeth), 196, 197
uterus, 80, 181, 182
uvula, 140, 141
valium, 345
valsalva manoeuvre, 133–34, 202
vasopressin, 341
varnish, 7
velopharyngeal insufficiency. See loss of seal
ventral horn (of spinal cord), 9, 10, 245
ventral roots, 8, 9
ventral striatum, 322, 323,
ventral tegmental area, 322, 323
ventricles: of the larynx, 140, 145, 146, 206;
of the brain, 218, 219, 222, 223, 229
vertebrae, 8, 17, 18, 19, 20, 21–26, 69, 109–
10, 112, 220
vertebral column. See spine
vestibular fold, 140, 145, 146, 206, 222
vestibular system, 219, 243, 292
vibrato in: singing, 124–25; flute playing,
124, 128–29; reed instruments, 133,
202; brass instruments, 202; stringed
instruments, 12, 93
vibrator (voice), 143, 144, 145–55
viola, 6, 8, 13, 24, 34, 36, 49–50, 77, 84, 92,
95; Erdesz design, 95, 96; “Pelligrini”
design, 95, 96
violin, 6, 8, 13, 24, 32, 34–36, 49–50, 84,
92, 247, 279, 293; design of, 95, 96,
[Animation 2-2 violin posture]
vital capacity, 103, 104, 106, 112, 120, 121,
122, 125, 126, 127, 129, 132–33
vocal cords. See vocal folds
vocal folds; breathing and, [Animation 5-2
larynx, breathing]; changes through life,
177–84, 187; chink between, 155, 182,
[Animation 5-3 larynx, voice quality];
in singing, 108, 114, 120, 124–25, 140,
144, 143, 145, 146, 147, 148, 149,
150–51, 152, 153–57, 158, 159–65,
170, 175, 182–84, 187, [Animation 5-3
larynx, voice quality], [Animation 5-4
larynx, pitch], [Animation 5-5 vocal fold
cycle], [Animation 5-6 vocal fold cycle,
continuous], [Video 5-1 normal larynx];
movements in wind playing, 129, 133–34,
202–3, 206; problems affecting, 172–80,
182, 183, 187, 189
vocal fry. See pulse register
vocal ligament, 146, 147, 148, 149, 150,
153, 157, 158, 159, 177, [Animation 5-
4 larynx, pitch], [Animation 5-7 vocal
timbre], [Video 5-1 normal larynx]
vocal misuse, 176–77
vocal nodules, 176, [Video 5-3 edema
nodules]
vocal polyps, 175
vocal tract, 113–14, 124–25, 143–73, 221;
resonances in singing. See formants;
resonances in wind playing, 202–3
vocal register, xiv, 144, 147, 153, 155, 156,
157, 158, 159–60, 179, [Animation 5-7
vocal timbre]
voice abuse, 176
voice: aging of, 177, 178, 179–80, 187;
breaking (see puberty, the voice and);
pitch regulation of, 143, 150, 153–55,
158, 159–60, [Animation 5-4 larynx,
pitch]
vowel intelligibility, singing and, 167, 168,
169, 170, 171
Wagner, Richard, 185
warm up, 75–76, 80, 86, 177
weight control, 342–44
Wernicke’s area. See cortex, areas of
whispering, 155, 174, [Animation 5-3 larynx,
voice quality]
white matter, 2, 9, 10, 16, 219, 222, 224,
249, 317
Williams, John, 37
windpipe. See trachea
Wolf, Virginia, 348
womb. See uterus
 Index 367

woodwind instruments, 7, 31, 73, 88, 92–93,
104, 128–29, 131, 133–35, 141, 196, 202,
207. Refer also to individual instruments
working memory, 311, 312, 315; failure of
in stress, 334
wrist posture, problems arising from, 39, 59,
78, 82, 88, 89, 90, 93, 94, 95, [Animation
2-3 piano posture], [Animation 3-4
retinacula of wrist], [Animation 3-5 wrist
position]
writer’s cramp. See also focal dystonia,
261–62, 265–66, 269
yoga, 40, 69, 73, 340, 342–43
zygomatic arch, 193, 197, 198
 About the Author

Alan Watson is a senior lecturer in anatomy and neuroscience in the School of Biosci-
ences at Cardiff University, and has had a lifelong interest in musical performance. He
was born in Kirkcaldy in the kingdom of Fife, Scotland, and educated at Glenrothes
High School. There he benefited from the excellent provision made for instrumental
music by the local education authority, learning both the French horn and the flute. He
took a B.Sc. in Zoology at the University of Edinburgh, and a Ph.D. in Neuroscience
at St. Andrews University before postdoctoral studies at Cambridge University, where
he held a Beit Memorial Fellowship for Medical Research. For the last twenty years
he has taught gross anatomy and neuroscience to medical and science students at Car-
diff University. He has published over sixty papers on neuroscience in professional
journals and worked on an eclectic range of subjects that resembles the contents of the
witches’ cauldron in Shakespeare’s Macbeth (eye of midge, gut of fish, leg of locust
. . . ). He has also dabbled in early keyboard and stringed instrument construction. In
Cardiff he runs a course on the biological principles underlying musical performance
for the Department of Lifelong Learning and also teaches on this subject at the Royal
Welsh College of Music and Drama. There he works with staff and students on proj-
ects concerned with understanding breathing patterns in wind players and singers
that are supported by the Wellcome Trust. He has also taught at the Royal College of
Music in London and has given public lectures at venues such as the DANA Centre
(London) and the British Association for Performing Arts Medicine, and at events
ranging from science festivals to the Menuhin Violin Competition.

369